Improving Stability in Developing Nations through Automation 2006 (IPV - IFAC Proceedings Volume) (IPV - IFAC Proceedings Volume)

59

Chung, S.-H., A.H.I., Lee, and W. L. Pearn (2005), Analytic network process (ANP) approach for product mix planning in semiconductor fabrication. Intl. J. of Production Economics, 96 (1): 15-36. Fredendall L.D, and B. R. Lea (1997), Improving the product mix heuristic in the theory of constraints. Intl. J. of Production Res., 35 (6): 1535-44. Finch, B. J., and R. L. Luebbe (2000), Response to 'Theory of constraints and linear programming: a re-examination'. Intl. J. of Production Research, 38 (6): 1465-66. Goldratt, E. M. (1990), What Is This Thing Called Theory of Constraints. New York: North River Press. Goldratt, E.M. (1993), What is the theory of constraints? APICS-The Performance Advantage, 3 (6): 18-20. Goguen, J. A. (1969), The logic of inexact concepts. Syntheses, 19:325-373. Hsu, T.-C, and Chung, S.-H. (1998), The TOC-based algorithm for solving product mix problems. Production Planning &Control, 9: 36-46. Kee, R. (2004), Evaluating product mix and capital budgeting decisions with an activity-based costing system. Advances in Management Accounting, 13: 77-98. Kee, R., and C. Schmidt (2000), A comparative analysis of utilizing activity-based costing and the theory of constraints for making product-mix decisions. Intl. J. of Production Economics, 63 (1): 1-17. Lea, B.-R., and L. D. Fredendall (2002), The impact of management accounting, product structure, product mix algorithm, and planning horizon on manufacturing performance. Intl. J. of Production Economics, 79 (3): 279-99. Lee, T. N., and G. Plenert (1993), Optimizing theory of constraints when new product alternatives exist. Product. & Invent. Management J., 34 (3): 51-57. Luebbe, R., and B. Finch (1992), Theory of constraints and linear programming: A comparison. Intl. J. of Production Res., 30 (6): 1471-1478. Morgan, L. O., and R. L. Daniels (2001), Integrating product mix and technology adoption decisions: a portfolio approach for evaluating advanced technologies in the automobile industry. J. of Operations Management, 19 (2): 219-38. Onwubolu, G. C , and M. Mutingi (2001), A genetic algorithm approach to the theory of constraints product mix problems. Production Planning & Control, 12 (1): 21-27 Plenart, G. (1993), Optimizing theory of constraints when multiple constrained resources exist. European J. Operat. Research, 70 (1): 126-33. Saaty, T.L. (1996), The Analytic Network Process. Pittsburgh, PA: RWS Publications, 1996. Souren, R., H. Ahn, and C. Schmitz (2005), Optimal product mix decisions based on the Theory of Constraints? Exposing rarely emphasized premises of throughput accounting. Intl. J. of Production Res., 43 (2): 361-74. Stephen, C. J. (2002), MATLAB Programming for Engineers. Australia: Brooks/Cole, 2002. Vasant, R., Nagarajan R., and S. Yaacob (2005). Fuzzy linear programming with vague objective coefficients in an uncertain environment. /. of the Operational Research

of fuzziness lesser total number of product-mix is obtained. Figure V demonstrates an interesting phenomenon also found in the nature: at lower cp the chances of getting involved higher degree of fuzziness (y) increase. Thus a DM's level of satisfaction should be at least moderate in order to avoid higher degree of fuzziness while making product-mix decision under TOC. Figure VI depicts a surface-cum-contour plot illustrating relationships among three prime parameters of product-mix decision. It reveals that the maximum and the minimum number of products to be produced, via utilizing resource capacity between 0% to 100%, are fixed. And the DM can adopt the intelligent decision ranging the said maximum and minimum values at his/her desired level of satisfaction.

3. CONCLUSIONS It was shown above that the fuzzy LP approach proposed for solving product-mix-selection problems outperforms considerably the earlier proposed approaches. The solution by Hsu and Chung via dominance-rule technique yields the optimal throughput as US $ 11,873, which is identical to the solution found by the present FLP methodology. Thus it is believed to be considerably better suited for this class of problems. The flexibility of the model lies in the membership function design. The present approach, in a nutshell, fuses two most important measures for the decision makers as well as implementers, into the existing TOCLP product-mix heuristic. In contrast, the genetic algorithm (GA) approach by Onwubolu and Mutingi (2001) fails to maximise the throughput. Moreover, even a cumbersome dominance rule (Hsu and Chung, 1998) is not required when the present approach is applied. Thus fusion of DM's expectations as well as vagueness of decision into a single model makes the fuzzified intelligent approach robust and more efficient to apply in TOC product-mix decision.

REFERENCES Anderson, S. W. (2001), Direct and indirect effects of product mix characteristics on capacity management decisions and operating performance. Intl. J. of Flexible Manufacturing Systems, 13 (3): 241-65. Bengtsson, J., and J. Olhager (2002), The impact of the product mix on the value of flexibility. Omega 2002; 30 (4): 265-73. Bhattacharya, A., B. Sarkar, and S. K. Mukherjee (2004), Use of DeNovo programming in theory of constraints (TOC). Industrial Engineering J., XXXIII (7): 6-11. Building GUI's with MATLAB version 7.0. The Math Works, Inc, USA.

60

Copyright © Improving Stability in Developing Nations through Automation Prishtina, Kosovo, 2006

ELSEVIER

IF AC PUBLICATIONS

LEARNMATHS: A CASE STUDY OF THE DEVELOPMENT OF LEARNING SOFTWARE TO SUPPORT SOCIAL INCLUSION M.A. Hersh1 and L. Stapleton^ Electronics and Electrical Engineering1, University of Glasgow, Glasgow G12 8LT, Scotland. Tel: +44 141 330 4906. Fax: +44 141 330 6004. Email: [email protected] ISOL Research Centre2' Waterford Institute of Technology, Waterford, Republic of Ireland Tel: +353-(0)51-302059, Fax: +353-(0)51-378292, Email: [email protected]

Abstract: This paper discusses the role of technology in overcoming existing barriers to education and employment faced by people with specific learning disabilities and illustrates the discussion by the example of the LearnMaths toolbox for people with specific learning disabilities in mathematics. The context of the toolbox is respect for equality and diversity and consequently recognition that different approaches to learning are equally valid and that the cognitive processes of people with specific learning disabilities can lead to creativity and innovation. Copyright© IF AC 2006. Keywords: mathematics, specific learning disability, assistive technology, user involvement, empowerment.

populations (Hersh, 2006), giving a need for an influx of younger people to work. However rather than welcoming asylum seekers, many of whom are of working age and have good qualifications, the richer and less unstable countries, which are a minority destination for asylum seekers (Hersh, 2006), detain and deport them. At the individual level, excluded and marginalised individuals experience poverty and discrimination, as well as diminished feelings of self-respect and self-worth.

1. BACKGROUND AND CONTEXT One of the main prerequisites for social stability is a stable environment in which people can live together in peace. This paper takes the view that social inclusion at the margins is a particularly important aspect of social stability. From the point of view of engineering researchers this means prioritising the needs of people who are marginalised by existing social structures, as well as involving them in the design process.

Diversity contributes to the wealth of society and there is even recognition that having a diverse workforce will contribute to a firm's profitability and competitiveness (Bertuglia, 1994). One aspect of diversity is differences in skills and abilities, learning strategies and cognitive processes, as well as areas of learning disability. There is increasing recognition of the learning disability dyslexia, despite some continuing controversy as to whether or not it really exists, and some progress has been made in developing learning approaches and technologies to support people with dyslexia. However less attention has been given to mathematics learning disabilities. Many everyday activities such as shopping, budgeting and decorating a room require basic

The term "social inclusion" is being increasingly used in European and other policy to express a belonging to and involvement in society and an ability to access goods and services and educational and employment opportunities. It is important because many people experience social exclusion and marginalisation due to a range of factors, including poverty, disability, sexual orientation, gender, ethnic origin and immigration status. This has costs to society as well as the individual, as people are forced into a position of dependency rather than being empowered to obtain education, training and employment and contribute to society. Many industrial societies have falling birthrights and ageing

61

numeracy skills. Adults who experience difficulties in telling the time or knowing what coins and notes they need to pay for something may feel inadequate and isolated, considering the fault to be in themselves rather than society. Numeracy skills also facilitate access to computers and the internet. More advanced mathematical skills are required for many careers and scientific and technical education at all levels. Particular difficulties in acquiring these skills are encountered by people with (specific) learning disabilities, such as dyscalculia and dyslexia, leading to potential disadvantage both with regards to everyday activities and education and career opportunities.

3. THE NEED FOR EDUCATION SUPPORT SOLUTIONS Changes in theories of learning from behaviourism to cognitive psychology and social constructivism have had a limited impact on the approaches used with students with specific learning disabilities. There has been a tendency to focus on behavioural approaches and rote learning of maths facts, rather than problem solving or imparting skills embedded in understanding (Woodward and Montague, 2002). Some curriculum aims may also unintentionally disadvantage people with learning disabilities. This has been discussed in the context of calculator use, where it has been suggested that present approaches, which often prohibit calculator use before arithmetic concepts have been assimilated (Orton-Flynn et al., 2000) should be modified to permit appropriate calculator use in combination with learning to estimate approximate answers. This would reduce short term memory demands, which can be a problem for people with specific learning disabilities, without reducing the need for understanding. Other issues relate to whether the currently used algorithms, which have evolved historically, are the most appropriate for students with learning disabilities or whether they would benefit from having the option to choose between several different algorithms (Woodward and Montague, 2002). Studies have also shown that the media chosen can affect learning outcomes and that there are advantages in a multisensory as well as a multi-media approach. However, multi-media and multi-sensory approaches need to be treated sensitively to avoid sensory overload (Beacham, 2002).

2. LEARNING DISABILITY IN MATHEMATICS An estimated 5-8% of school-age children have specific learning disabilities in one or more areas of mathematics (Geary, 2004). Empirical studies have found that children with specific learning difficulties in mathematics often use strategies which do not require mental retrieval of number facts (Gersten et al., 2005). This supports theories that they have poor short term memory retrieval skills (Dowker, 2005). There is also empirical evidence that one of the main distinguishing features of children with dyscalculia is difficulties with multistep arithmetic (Bryant et al., 2000). Since the term mathematics encompasses diverse topics which make different cognitive demands, specific learning disabilities may impact on some areas of mathematics, but not others (Geary, 2004). However most research in the area has concentrated on arithmetic and number skills. A relatively small number of both computer-based and other tools have been developed to support people with (specific) learning difficulties. However most of the commercially or freely available tools focus on children, rather than adults or the full age range. A study of the literature indicates that coverage of mathematics topics is also relatively limited and generally related to basic numeracy skills rather than more advanced topics and concepts.

There has been some progress in understanding the learning processes of people with (specific) learning difficulties. However, a taxonomy of mathematical learning difficulties has not yet been developed. Neither has a categorisation of the different areas of mathematics, correlated to the different types of mathematical learning difficulties. Consequently, the assistive technologies available to people requiring support in this area tend to be narrowly focussed and relatively inflexible. What is needed is a comprehensive assistive technology solution which provides a coherent set of tools for the user. The author's call this the 'LearnMaths Toolkit'.

There is therefore a need for further research on tools, methodologies and learning approaches to support mathematical and numeracy learning by people with dyscalculia of all ages and stages of education. There are also questions as to the role of (computer) technology in this process, whether and in what circumstances it can be an enabler rather than a further hurdle to be overcome and the appropriate balance between the computer based resources and other types of support. The authors attempt to address the following proposition: Proposition: is it possible to design a support system which can enable people who experience social exclusion as a result of mathematical learning difficulties?

3.1 End- User Involvement The Learnmaths Toolbox and Learning Environment is intended to support students with dyscalculia in acquiring skills and understanding from basic numeracy to advanced mathematical techniques. Since it is only end-users of a system who really understand their own needs and what is likely to work for them, the content of the toolbox will be finalised after consultation with end-users with dyscalculia.

62

End-user involvement will be organised according to the principles of human centred technological systems development and the principles developed in the EU FORTUNE Project (Biihler, 2000). The FORTUNE concept of user participation is based on the following principles: 1. Partnership as a basis. 2. Users as members and/or representatives of organisations. 3. Users should receive payment on the same basis as other participants. 4. Guaranteed accessibility of all relevant materials and premises. 5. All partners guarantee confidentiality, respect and expertise. 6. A detailed project plan, including time and resource planning for user participation. 7. Partnership is implemented from the start of the project.

'better' and increased effectiveness may be the consequence of a social context which privileges some approaches over others. Individuals who have cognitive and learning approaches which are not privileged are likely to experience disadvantage and discrimination. To avoid this educational tools are required which provide different approaches and aim to communicate with users on their terms rather than trying to mould them to meet norms. This will have some effect on changing attitudes so that different is not considered less or socially unacceptable. This is particularly important in avoiding stigma, low self esteem and feelings of alienation, as well as overcoming barriers to students with specific learning (or other) disabilities accessing the resources and teaching support they require. However full discussion of changing attitudes is beyond the scope of this paper. The constructivist approach to learning is based on the premise that learning is a by-product of understanding, which is situated in a personal, social and organisational context, requires frequent feedback from teachers and fellow students and the performance of tasks (Mayes and Fowler, 1999). It has also been suggested that conceptual learning depends on reflection and making information meaningful with regards to existing cognitive structures and with sufficient richness and elaboration (Craik and Tulving, 1975). This pedagogical philosophy confirms the importance of taking account of cognitive diversity and providing a structure which is compatible with the experiences and cognitive approaches of the learner with specific learning disabilities. Rather than being a discrete activity, conceptual learning is probably a cycle involving a process of continuous development and testing of understanding against internally or externally defined criteria (Mayes and Fowler, 1999). It should be noted that this cycle will not necessarily always involve a continuous increase in understanding, but may also comprise misunderstandings and their correction.

This will be combined with a human centred approach based on the H-CAD (Human Centred Assistive Technology systems Development) methodology (Duffy et al., 2004; Stapleton et al., 2004). This methodology has been applied successfully to the design and implementation of assistive systems for adult learners with dyslexia. It has been shown with adult learners with dyslexia in Ireland that the use of a human centred approach reduces the likelihood of the learners having increased feelings of isolation. This can often be a problem in assistive technology projects for people with specific learning difficulties. The H-CAD methodology significantly reduces the likelihood of this problem by careful management of user involvement and contextualisation of the work.

3.2 LearnMaths Difference

Philosophy:

Diversity

and

The philosophy of the LearnMaths Toolbox is based on the social model of disability and respect for equality and diversity. According to the social model of disability it is social barriers and oppression which disable people rather than their impairments. For instance, it is the steps not the wheelchair or the lack of large print books not the visual impairment which are the problem. Therefore it is society rather than disabled people which needs to be 'rehabilitated'. A philosophy based on equality and diversity recognises both the importance of diversity or difference and the entitlement to equal rights and opportunities. However, access to these equal rights and opportunities may be obtained by different groups of people or individuals in different ways.

In the university context, despite increasing commodification of both teaching and research and the tendency to suppress minority opinions and (Hersh and Moss, 2004), there is still recognition of the value of diversity, including of the subjects covered and the academic community. There is a need for this recognition of the value of diversity to be extended to the other levels of education and the wider society. This will enable the full diversity of the population, including disabled people with a variety of cognitive approaches, to participate and contribute to society and society to benefit from their different cognitive approaches. This could have significant benefits in terms of creative and currently unexpected solutions to existing social, political and other problems, as well as the development of technology and social systems which significantly contribute to improving the quality of life for the

In the context of learning and learning disability this implies that there are different approaches to cognition, organising information and learning. Some approaches may seem to be more effective than others, but no approaches are intrinsically

63

whole population. However, this will require educational and pedagogical strategies which enable people with different cognitive approaches, including specific learning disabilities, to obtain a good education and develop their creativity, rather than leading to their failure, as is generally the case at present.

different resources and approaches should be available to all students. For instance a study of the provision of numeracy materials to Open University students found that 93% of them preferred having access to both computer added learning materials and text based materials (Jones et al., 1999). These conclusions can probably be generalised to other groups of students, including those with specific learning disabilities, as long as the number of different types of materials to choose from is not so large that it imposes a cognitive burden. Provision of sufficient resources by the institution, including teaching materials and staff time (academic, technician, secretarial, administrative and other) to prepare the different approaches and resources. This is unfortunately counter to current trends in countries such as the UK, where the staff: student ratio has increased from 1:9 in the mid-1970s to to nearly 1:21 in 2003-4 and is expected to increase to 1:23 by 2010 (AUT, undated, 2002). The use of creative approaches. The involvement of students in course design, and flexibility in taking account of the needs, circumstances and context of all students, rather than a rigid application of principles.

3.3 Pedagogy of Diversity There is, therefore, a need for a pedagogy of diversity, based on appropriate learning methodologies and approaches for the full diversity of students and which enable all students to draw on their strengths. This pedagogy would need to include consideration of the following factors: • The essential learning goals of the particular educational programme. Often an educational programme may have a number of goals which are not really essential and which have developed for historical or other reasons. Removing these goals may make the programme accessible to disabled people. • Learning approaches which encourage students to meet the learning objectives while taking into account different learning styles. This is particularly important for students with specific learning disabilities. • Social aspects of education, including the need for peer acceptance. In some cases this will mean modifying the basic approach to make it accessible to all students. For instance one researcher suggests that the same educational goals could be covered by offering students with mobility impairments a field trip that does not require climbing mountains (Phipps and Kelly, 2006). However, this raises the issue that it would probably be preferable to either offer all the students the same field trip which does not require mountains to be climbed. Alternatively different groups of students could undertake one of a number of field trip options and share their experiences afterwards. • Access to the same types of experience, such as field trips and laboratory work. • Support for student learning by encouraging and providing approaches which enable all students to draw on their strengths and to use these strengths to support the areas in which they have weaknesses (Price, 2006). • Challenging and stretching all students to the optimal extent for the particular student, without making any student feel overwhelmed or stressed. • Provision of assignment measures and methodologies which are fair to all students. • Provision of a number of different learning and teaching approaches and a range of resources, with support to help students to choose the most appropriate approaches and resources. All these

3.4 Accessibility and Usability In the context of technology for disabled people there has been a tendency to focus on accessibility, for instance, of computer system or web pages. Accessibility is about the environmental characteristics of the system input and output which either enable or prevent particular groups of users from using the system, whereas usability is the ability of the system to carry out the intended function(s) when used by particular groups of users (Federici et al., 2005). It has been defined as the extent to which a particular group of users can use a particular system in their particular context to achieve specified goals effectively, efficiently and with satisfaction (ISO, 1998). Accessibility and usability are both essential for all users, since users need to both be able to access the system (input to it and receive outputs from it in a comprehensible form) and use it to carry out the intended functions. However educational software, for instance, may need to challenge the learner to make them think. This could require the software to be designed to avoid too much seamless fluency in use, as this may prevent deep learning (Mayes and Fowler, 1999). On the other hand excessive software complexity and cognitve demand should also be avoided. Approaches to usability include characterising it in terms of the following six components (McLaughlin and Skinner, 2000):

64

•

• •

• • •

4.2 Benefits of the LearnMaths Toolbox

Checkability: allowing checks to ensure that the correct information is entering and leaving the system. Confidence: users are confident in both the system and their ability to use it. Control: users have control over system operations, including input and output information. Ease of use: the system is easy (and intuitive) to use. Speed: the system responds and carries out functions quickly. Understanding: users are able to understand the system and its outputs.

The development of the Toolbox will take place in the context of research on how people with learning disabilities learn with regards to both understanding concepts and the acquisition of skills. This could have a wide range of benefits with regards to improving understanding of cognitive processes and the particular skills of people with dyscalculia. However, the main benefits will be improving the opportunities of people with dyscalculia and dyslexia in education and employment, increasing their autonomy and encouraging their participation in and contribution to society. Improved numeracy skills will lead to independence and autonomy in areas such as budgeting and shopping. People with specific learning difficulties in mathematics, who are an estimated 5-8% of the population will benefit in the following ways:

Both usability and accessibility should be considered as part of good design practice and included from the earliest stages of system specification and design. Consideration of usability and accessibility also require involvement of end users in all stages of research and development, as discussed in section 3.1.

• • •

4. THE LEARNMATHS ASSITIVE TECHNOLOGY SOLUTION 4.1 The LearnMaths Toolkit The initial version of the Toolbox will focus on tools for the 16+ age range, with further work to develop and extend the tools to younger people. The Toolbox will cover a wide range of different mathematical and numeracy skills. Some of the tools will be specific to particular areas of mathematics, whereas others will be more general. The Toolbox will also provide a multimodal, multi-sensory and multi-media capability, including a choice of output modes and the option to configure the system to meet user preferences. This is in-line with research evidence that people with specific learning disabilities can benefit from multi-sensory and multi-media capacity, but that care is required to avoid sensory overload.

•

•

Improved educational outcomes and qualifications. An increased ability to participate in the Information Society. An increased ability to participate in policy debates about technology. As the role of technology increases, the general public will require a better scientific education in order to understand the issues. It is important that people with learning difficulties are not left behind. An increase in self-esteem and more enjoyable educational experiences as a result of appropriate strategies being used to avoid (repeated) failure. Access to a wider range of (professional) careers, including in the sciences and engineering (for people with milder mathematical learning difficulties.

There will also be benefits to the following groups of people: • Teachers of numeracy skills, who will benefit from additional tools to support their work. • Learning support tutors, particularly those supporting students with dyscalculia and dyslexia. • Schools, colleges and universities, which will benefit from increased success rates in mathematics and subjects dependent on it. • Society as a whole, which will benefit from the increased contribution made by people with dyscalculia and dyslexia.

The customisation facility will enable users to choose the optimum learning environment to meet their needs, including when they are fresh and when they are tired. Attention will be given to making all the LearnMaths tools fully accessible, so that they can be used by students with additional impairments as well as dyscalculia. The Toolbox may also be of interest to a number of students who do not have any recognized learning disabilities, but who encounter difficulties with current learning and teaching approaches or who find them uninteresting. Unlike many existing assistive systems, the Toolbox will be designed to support both the acquisition of practical skills and the development of understanding of basic concepts.

Although the main aim of the LearnMaths Toolbox is to support people with dyscalculia and dyslexia, the multi-sensory and accessible nature of the tools in the Toolbox will mean that many of the tools will also be useful to blind and visually impaired people. Therefore there will also be subsidiary benefits in terms of improving the opportunities of blind and visually impaired people in the areas of education

65

and employment, increasing their autonomy and encouraging their participation in and contribution to society.

5.

Craik, F.I.M. and E. Tulving (1975). Depth of processing and the retention of words in episodic memory, J. of Experimental Psychology, 104, 268-294. Dowker, A. (2005). Early identification and intervention for students with mathematics difficulties, J. Learning Disabilities, 38(4), 324332. Duffy, D., L. Stapleton, M. Jordanova, D. Lakov and M. Lyng (2004). From assistive technologies to assistive systems: total solutions for the learning disabled, Proc. MEDeTEL Int. Conf Luxembourg. Federici, S. et al. (2005). Checking an integrated model of web accessibility and usability evaluation for disabled people, Disability and Rehabilitation, 27(13), 781-790. Geary, D.C. (2004). Mathematics and learning disabilities. J. Learning Disabilities, 37(1), 4-15. Gersten, R., N.C. Jordan and Flojo (2005). Early identification and intervention for students with mathematics difficulties, J. of Learning Disabilities, vol. 38(4), 293-204. Hersh, M.A. and G. Moss (2004). Heresy and , Int. J. of Gender Studies. Hersh, M.A. (2006). Mathematical Modelling for Sustainable Development, Springer Verlag. ISO (1998). Ergonomic Requirements for Office Work with Visual Display Terminals (VDTs)— Part 11: Guidance on Usability 9241-11, International Organization for Standardization. Jones, A et al. (1999). Contexts for evaluating educational software, Interacting with Computers, 11,499-516. McLaughlin, J. and D. Skinner (2000). Developing usability and utility: a comparative study of the users of new IT, Technology Analysis and Strategic Management, 12(3), 413-423. Mayes, J.T. and C.J. Fowler (1999). Learning technology and usability: a framework for understanding courseware, Interacting with Computers, 11,485-497. Orton-Flynn, S and C.J. Richards (2000). The design and evaluation of an interactive calculator for children, Digital Creativity, 11(4), 205-217. Phipps, L. and B. Kelly (2006). Holistic approaches to e-learning accessibility, ALT-J, Research in Learning Technology, 14(1), 69-78. Price, G.A. (2006). Creative solutions to making the technology work: three case studies of dyslexic writers in higher education, Research in Learning Technology, 14(1), 21-38. Stapleton, L., D. Duffy, M. Jordanova, D. Lakov and M. Lyng (2004). From assistive technologies to assistive systems: human centred support for the learning disabled, Int. Multitrack Conf. of Advances in Control Systems, TU Wien. Woodward, J. and M. Montague (2002). Meeting the challenge of mathematics reform for students withLD. J. of Special Education, 36(2), 89-101

CONCLUSIONS

The paper has discussed the role of technology in overcoming the existing barriers to education and employment faced by people with learning disabilities and presented the LearnMaths Toolbox as an example of a technological assistive system that can be used to provide support in overcoming these barriers. The discussion has been in the context of equality and diversity, with equal value given to different cognitive styles and learning approaches. This is particularly important for people with specific learning disabilities in preventing the disadvantages, discrimination and exclusion they are otherwise likely to experience. The discussion of equality and diversity has led to the presentation of some of the underlying principles of a pedagogy of diversity. The LearnMaths Toolbox has also been used to illustrate the importance of involving end-users in all stages of the development of technology and to give examples of some of the approaches to doing this. This has led to a clarification of the differences between accessibility and usability and a discussion of the complementary importance of both properties in educational software. Finally the paper demonstrates a technical solution to social stability by providing an assistive system for people marginalised by society's lack of flexibility with regards to different approaches to learning and different learning needs.

REFERENCES AUT (undated), Staff-Student Ratios, http://www.aut.org.Uk/index.cfm/media/pdf/c/3/m edia/pdf/index.cfm?articleid=l 33. AUT (2002), Nation's Universities in Crisis, says AUT, http://www.aut.org.uk/index.cfm/media/ pdf/c/3/media/pdf/index.cfm?articleid=290. Beacham, N., (2002). Dyslexia-friendly computerbased learning materials. In: Access All Areas: Disability, Technology and Learning, Phipps, L., Sutherland, A. and Seale J. (eds.), JISC TechDis Service and Alt. Bertuglia, L.E. (1994). An exploration of the glass ceiling, Women Eng., May/June, 27-29. Biihler, C. (2000). FORTUNE Guide Empowered Participation of Users with Disabilities in Projects, Evangelische Stiftun Volmarstein. Bryant, D., B.R. Bryant and D.D. Hammill (2000). Characteristic behaviors of students with learning disabilities who have teacher-defined math weakness. J. Learning Disabilities, 33, 168-179.

66

Copyright © Improving Stability in Developing Nations through Automation Prishtina, Kosovo, 2006

ELSEVIER

IF AC PUBLICATIONS

CURRENT ETHICAL SCRUPLES CONCERNING INFORMATION ENGINEERING

Tibor Vamos

Computer and Automation Research Institute, Hungarian Academy of Sciences H-llll Budapest, Ldgymdnyosi u. 11., Hungary vamos(q)sztaki.hu

Abstract: In the irresistible emergence of e-society the individual is represented by his/her electronic identity. The theoretical and practical pros and cons of electronic identity are treated, the harmonization and contradictions of privacy and security interests, and their relationship to the principles of liberty and solidarity, individual and society. The issues of democratic control and influence to further developments of democracy are discussed with a special focus on the ethical duties and possible actions of the engineering community, with some references to the dramatic events of global terrorism, extreme fundamentalism and to the local efforts of Hungarian Information technology. Copyright@2006 IF AC Keywords: Social and behavioural sciences, social impact, linguistic support

wisdom. The second has been a dispute for two and a half millennia, nevertheless in a very new context, the first is brand new and did not really dominate relations before the Age of Reason, i.e. about a quarter of century ago, though its' emergence in everyday life of everyday people started no earlier than in the present times of the information society.

1. INTRODUCTION Through out the ages, no responsible professional activity has existed without scruples. Though, in our age unscrupulous behaviour is, in technologically advanced countries, less of a danger than it used to be earlier, even in the 20th century. The other addendum is the requisite of the concept of professional responsibility; the quality of being scrupulous, scrupulous in his/her science, personal competence and especially in considering the consequences of his/her results. The characteristic difference in our age is the ubiquity of science and science-originated technology in every relation of human life and the special role of information, disseminated and transmitted globally, as the fundamental vehicle in human action and society organization. This entails a global responsibility, requiring the deepest insight, foresight in distance and time in a world of minimal access distances and time delays with a maximum effectiveness of action.

2. PRIVACY AND SECURITY, TWO DRAMAS OF OUR DAYS Even the concept of privacy is new and so, therefore, is its close relation with security. The emergence of the modern definition started with information technology and democracy development in the US with the sentence of Associate Justice Brandeis: the right to be let alone. The change can be estimated after the relatively recent British imperial rule for the queen: she had to give birth to a child in the presence of state dignitaries. The Brandeis-idea was the result of the proliferation of newspapers and photoportraits.

In this paper two crucial problems will be outlined. The first is the combined issue of privacy and security; the other is even more bound to the contradictions of democracy and governance

The present, acute problem is due to the global menace of terrorism, the milestone: September 11. The other side of the recent picture is mirrored in the

67

global hysteria surrounding a caricature in a minor Danish satire magazine. The related problems are: What are the special messages to be issued for professionals working in information technology? Can a global, general and durable set of principles be recommended for the freedom of the individual doing what she/he considers to be un-harmful to others?

methods for codes and the various legal-procedural ways of security system control. This message is an extremely important one; it is the first answer to our question about our special message. This security system could have been a safety measure against terrorism and is now an efficient way of security control for private business and for sensitive information about planned malevolent actions. The privacy side of the system is a firewall of private information within the framework of legal action. The system can also work in defence of group privacy if it is agreed within the group and among the groups concerned. A rather welcomed but pragmatic agreement was compromised between Google and China. Some similar ways of negotiations can save the fundamental principles of our freedom of opinion and that of the Muslim faith.

What can, and what cannot, these people do for a more human mankind? Can they sketch a unique and realistic picture of that ideal, or, to express themselves in a more modest manner, a somehow better world? What does it mean to be better or ideal, or even realistically ideal? Can that realistic approach be exemplified by alternative scripts in discussing the two dramas of the introduction?

If this message is distributed well, i.e. in appropriate ways for understanding by people educated in the schools of our civilization, people can be prepared for the next, even more complex problems.

How can a balance of compromises be suggested by using our ethical principles?

4. WHO GUARDS THE GUARDS? — FEEDBACK OF ENLIGHTENED DEMOCRACY

3. PRIVACY AND SECURITY MEANS LIBERTY AND SOLIDARITY WITH ALL THEIR COHERENCE AND CONTRADICTIONS

The first is the supervision of these systems, the Latin question: quis custodiat ipsos custodies, who guards the guards? The second is fundamental problem of all human interactions: cases of conflict of interest.

The key problem related to our profession is the privacy, and/or security, issue that can be converted into liberty and common interest or into a liberty and solidarity problem. In this view, to be explained, the fundamental questions are raised by the quoted examples. For discussion, another extension of the usual definitions can be used: privacy should be understood in the usual frame of the basic individual right but also in the right of any cultural entity, e.g. religion, other beliefs, national and group way of life.

old i.e. the the

The answers for the first are also old, but insufficient. The division of power, the methods of escrow of keys are well practiced in traditional democracies and have no real effect in weak democracies and dictatorships, even if they are deceptive as was the case in several paragraphs of the Soviet Constitution. The only final answer can be the democracy of self conscious, rational thinking, enlightened people. This is the other message and in this attributed form. The last period's experience demonstrated the dangers of virtual democracy when introduced in an environment with no democratic traditions. Even in periods and countries infected by irrational emotional influences democracy can be misleading. According to the other aspect, and this is more close to my one, democracy is a historical process of mass practice and this practice must be a real exercise, however painful.

Security is understood in this view in the same individual and extended social community related context. The extension of the privacy concept towards a community cultural identity is much older than the privacy idea itself. The two meanings were mostly fighting against each other, the modern libertarian view declared the priority of the individual, and all ideologies claimed the contrary. In our view, now, it should be the continuous task of rational conflict resolution. This will be discussed a little bit more in details, later. In this view liberty is the security against any kind of physical and mental intrusion into the sphere of privacy, the act of being left alone by any other person or group. It looks to be harmonious if there is no conflict of interest.

The special message in the role of information technology is the practice of feedback. Feedback control is not only a basic quality of physical systems but generally valid for all kinds of system control in stable and optimal conditions. The instant possibility of direct, individual feedback can be the best warrant for maintaining individual security against all kinds of intrusion efforts from the very start. Principiis obsta, sero medicina paratur, resist at the beginning, late medicament does not work, this wisdom of Ovid

The harmony of privacy and security is, for the first time in history, given in exchange of information by the last two decade's progress in cryptography and algorithmic theory, by the asymmetric coding idea of the public key, by the random number generating

68

is not only valid for love but for every more harmful emotion, like nationalism, fundamentalist beliefs, racism, and adoration of political idols.

Let us return to the essential subject: ethics. The relativity of ethics and the ideal human conduct can be followed throughout the course of history. This should be an argument for teaching history and humanistic knowledge in the school, learning from our own, multi-generation experiences. Even for those who preach the Bible as being the eternal stone tablet with its ten commandments have scope for interpretation and reinterpretation in respect of more peaceful timer.

The best example of continuous control by electronic feedback information is widely given by the bank account systems with possibilities for instant checking of any agreed kind and limit of personal and alien operations. The same practice should be regular related to any intervention in personal data, in operations of the community. The freedom of public information, its control and the privacy of individual data are on line controllable and any deviation can be identified at the very start. No democratic control in history has had such a power and it can now be introduced into a conscious, democratically, trained population. Switzerland, the most libertarian country in Europe now introduces broadband access for each citizen as a human right. Education, in how it can be appropriately used, would advisably be taught firstly in the primary school.

Unfortunately, even computer scientists are not empowered by our knowledge to speak on the subject of final truth and long lasting milestones. Nevertheless, some advice and principles can be suggested: With the conviction gained from complexity and computability theory it can be said that no deep and broadly manifest social problem can be solved in an ideal way. Each situation should be analysed in an unbiased fashion regarding all pros and cons and our duty is to proclaim, teach and, if possible, apply the least harmful, most effective solution. This should always be free for revision if the reality speaks against the hypotheses. This also means a determined attitude and work ethic in research, design and experimentation.

5. ADVANCES IN CONFLICT RESOLUTION The next message of our science is the progress in conflict resolution. Philosophy concerning human nature, coexistence and organizational, ruling methods are rooted in ancient Greek wisdom, much deeper than in any earlier or subsequent beliefs. Modelling and the analysis of alternatives continued in the Age of Reason, but mathematical, computational treatment is plausible only with our current technology. Game theory combined with statistics and the statistical observations of psychologists are all achievements of the last halfcentury and honoured highly with a sequence of Nobel Prize Awards. The methods of iterative, nonzero, multiagent games, negotiations with quasioptimal, multivariate equilibrium and calculations with uncertain information about the state, and the intentions of the participants, have grown to a powerful weapon in delicate international affairs, e.g. in the Cuban rocket crisis or later in several reported, and unreported, conflicts endangering mankind. It cannot be said that the principle of rational choice has become a general tool in the conduct of life, not even that these scientific, computational methods were applied or could be applied and applied successfully in every great decision process but, as it was emphasized about the practice of democracy, the methodology and practice is a long and necessary avenue of international relations; The trend is encouraging and this promising issue has to be followed and promulgated as a special role in information and control of scientists.

This is what our world wide does community currently in search of an optimal design of information societies. The information society is mostly an artefact, much more than any previous society formations. An artefact, which is a result of science. Our special responsibility in shaping that society is therefore exceptional. The preservation of privacy and security is the key social and information science related task. The ethics of this sceptic view is a strong one. It cannot rely on ready-made principles given by any higher authorities, on an ongoing basis our knowledge, views, feedback and mental strength have to be investigated in a persistent but modest way.

7. HUNGARIAN SITUATION AND SOME ABOUT OUR GROUP'S ACTIVITY Our group is active in the field, too. The speciality of the Hungarian situation is characterized by, a historically imprinted suspicion against any state power, a direct result of the state being subject to foreign intervention for long periods in its history. However, Hungary was rarely subjected in such crude ways that could generate a firm revolutionism. Cohesion was created in rather short periods in 1703, 1848 and 1956 as these semi-independency situations looked to be changeable after the intervention years. This semi-independence status at first created an advantage but, subsequently, something of

6. STRONG ETHICS OF A SCEPTIC VIEW

69

reluctance. This is the reason why Hungary is now in the middle range of the new ten EU members and is not as pioneering as it was during the last quarter of a century before the changes.

All measures which are planned, and in progress in the EU and within, Hungary are controlled by the principles of maximum possible voluntary introduction. The general application of electronics in our social-individual activity depends on its attractiveness, economy and safely. On the other hand, the task is urgent in the framework of global competition and in view of emerging malevolent forces. Science and engineering wisdom should balance and support the right measures.

The major task of our academic group now is to help the introduction of an EU standard system for the above objectives, taking into account several local peculiarities, slow standardization and the fact that the EU that was unable to reach a consensus on the Constitution and incorporating all this into the framework of a rather radical innovation of public service systems and operations. The latter is, according to our views, an integral condition of a working information society.

8. CONCLUSION AND WARNING The progress of technology creates both progress and problematic side effects in society. The relevance and meaning of the individual and her/his rights has never been as clear and emphasized as in the information age. Science and technology offer splendid vehicles for this social progress; these are briefly discussed in the paper with references to the harmony and contradictions of privacy, security, freedom and solidarity, all of which, in a broad, modern sense, are the main values of our civilization. Nevertheless, a warning should be added. Malevolence is always at the edge of goodwill with its evil absence of scruples and ethical constraints and, that is also valid to our subject.

Education, and enlightening information via the mass media and other methods, is a necessary part of this. A further consideration in the longer term is that a modern information society administration should create a man-machine dialogue with the citizen, supporting the focus on those problems that need special human care and judgment. All other communication is planned to be perfected in natural language, i. e. natural language understanding is one of the ongoing researches, with a special emphasis on understanding low quality texts for relatively uneducated citizens, understanding of vocal communication and automation of administrative reasoning to the frontiers of the above limitations. These limitations are never fixed, they depend on the state of art of cognitive, computer, social and legal science. Recognizing the current limits is a major research goal of the group consisting of specialists in these fields, with linguists and enthusiasts of administrative practice. A report on the status of this project is planned for the current year's European Artificial Intelligence Conference.

For progress there is no cure as was said by Neumann, half a century ago and, he added, all menaces also present in our current times. Anybody should be aware of the fact that no final panacea exists against ingenuity and unrestraint of evil forces and that obliges our engineering society to serve progress in a positive, enlightening way.

70

Copyright © Improving Stability in Developing Nations through Automation Prishtina, Kosovo, 2006

ELSEVIER

IF AC PUBLICATIONS

COMPLEXITY OF SOCIAL SYSTEM CYBERNETICS: RISK AND UNCERTAINTY MANAGEMENT IN NEGOTIATIONS M. J. Stankovski1, T. D. Kolemisevska-Gugulovska1, Z. A. Icev1, A. T. Dinibtitun2, and G. M. Dimirovski1'2 1

SS Cyril and Methodius University, Faculty of Electrical Engineering - IASE Karpos 2 bb, MK-1000 Skopje, Rep. of Macedonia [email protected]; {milestk; tanjakg; z_icev} @etfukim.edu.mk Dogus University, Faculty of Engineering - DCE Acibadem-Kadikoy, Zeamet Sk21, TR-3 4722 Istanbul, Rep. of Turkey {tlahad; [email protected]. tr

Abstract: The globalization era world-wide involve mass communications and competition of conflicting interests, social and society interactions as well as various confrontations intrinsic to which are the negotiation processes based on information and decision analyses. This social system cybernetics emphasises for the mankind society at large the improvement society's ability to manage uncertainty has become vitally important. Probability theory and information theory as well as fuzzy-set theory, and thereafter info-gap and risk-management models have contributed considerably to understanding social system cybernetics as well as an managing international conflicts and risks. By adopting the information as the third fundamental category next to those of energy and matter, and taking theories of complexity and complex systems as point of departure, it can be argued in favour of composite linguistic and probabilistic information as categorical concept in negotiations along with uncertainty management within organizational human-centred and social systems. Copyright © 2006 IF AC Keywords: Decision making, human centred systems, information, negotiation, uncertainty.

1. INTRODUCTION The social impact of automation on the modern civilization of mankind has been enormous during the past decades along with certain unexpected developments. Wherever world-wide political forums have taken place, there is widespread anti-globalization protest with sometimes dramatic international participation. This is internationalised negative civic reaction is getting more organised and rumours about a kind of world-wide anti-globalisation movement spread. Yet, it may be argued the current process of world-wide interaction and integration of economies as well as the other life-sustaining activities of contemporary societies (Glenn and Hannan, 2000; Pete et al, 1998, Petit, 1990) actually is the inevitable stage of mankind's historical development (Dimirovski et al, 2006).

Also it may be argued that, 15 years after the end of the so-called bipolar world (Brams, 1994; Howard, 1971) during the Cold War era, the world-wide general public awareness is shifting towards a critical perception and re-assessment of modern civilization as a unipolar world (Eeckhout, 2000; Haime, 1998; Royal Society, 1992; Wagner, 1994). For, apparently endangered by certain negative global consequences such as environmental pollution, global warming, the so-called "north-south gap", and terrorism warfare, the mankind societies cannot move forward safely (Kile, 2005; Dinibutun and Dimirovski, 2001; Kopacek and Stapleton, 2004). Mansour, 2001). Moreover even some kind of global dictatorship, the systemic structure of which remarkably resembles the known such events in developed countries not so long ago, cannot be excluded.

71

The famous mathematician, philosopher, and humanist Bertrand Russell (1975) has pointed out, on the other hand, a fundamental concept in human societies as well as in human-centred and social systems is power with its many forms: civil authority, clergy, public propaganda, secret service, and the military, and - above all - the wealth. This fact and the need for competition in human societies as well as other factors of western social philosophies, largely adopted world-wide, have been inducing potential for conflict. Thus these have been producing and re-producing social system behaviour resembling the non-cooperative games (Basar and Olsder, 1999; Kreps, 1995), and often the conflict noncooperative games (Osborne and Rubinstein, 1994). In addition, generally science and technology, hence computing and control in particular, have a crucial impact on all these forms of power. On the other hand, all the forms of power have been and are being generated via some kind of organizational system (Gibson et al., 1997; Glenn and Hannan, 2000; Pete, 1998) which may imply further consequences both positive as well as negative. In this work, it is argued that such a growing internationalised and negative civic reaction, already taking place, to a great extent is due to the inappropriate uncertainty management accompanying all decision and negotiation processes. Crucially, the limitations of the existing approaches, either probabilistic (Ayers, 1994; Haimes, 1998; Funtovicz and Ravetz, 1990) or fuzzy-set possibilistic (Klir, 1997; Kosko, 1993; Zimmerman, 1997), still prevent these to cope with uncertainly management in social system negotiation processes. Despite the largely non-cooperative and/or conflicting cybernetics of social systems, in order the philosophy of Nash equilibrium (Nash, 1951) to prevail within the cybernetics of contemporary community of mankind societies, the societies at large must improve the ability to manage uncertainty in negations taking place among and within social systems. An innovated understanding of the information and the accompanying uncertainly, along with proper understanding of their composition facets as well as various computer processing techniques and application tools are urgently needed.

2. NEGOTIATION STRUCTURE: ISSUES AND VIEWS ON INFORMATION AND UNCERTAINTY

supervision applications which underlie the policy making in all human centred systems. Notice the statement by Funtowicz and Ravetz (1990): "Policies can no longer be assumed to be based on scientific information that is endowed with a high degree of certainty". For, the existing definitions of uncertainty in scientific information are typically very specific and very operation oriented as well as befitted to practically implementable tools for processing. Economics and financial engineering, risk assessment and postoptimization sensitivity analysis, political and postnormal sciences, and let alone applied mathematics and systems science; they all devote a great deal of attention to uncertainty. It appears, this concept is still not deeply understood hence defined poorly. True, despite this lack of common standpoint, a number of well-developed bodies of theory claiming to deal efficiently with the management of uncertainty exist: probability theory (Haimes, 1998) in the first place; then information theory (Krippendorf, 1986; Mansour, 2001); fuzzy sets (Klir, 1997; Zadeh, 1973; Zimmermann, 1997); and info-gap modelling (Ben-Haim, 1997). However, each of the techniques based on these concepts do explain only the kind of uncertainty which it deals with. Therefore the quest for common theoretical grounds continues and, as pointed out in Natke and Ben-Haim (1997), Klir as well as BenHaim argue that probability theory is synonymous with uncertainty theory to the majority of decision-making practitioners in various fields. Fairly recently Hatfield and Hipel (1999) revisited the issue of concern and proposed, via placing knowledge, uncertainty, and information into a comparative framework, to take the information as the fundamental concept and one these grounds redefine the other concepts involved. However, this viewpoint on the information as the fundamental category concept has been put forward quite some time ago in (Dimirovski et al, 1977; Gitt, 1987; MacFarlane, 1993) and also related to knowledge hence uncertainly too. Now it is re-examined in the light of the arguments presented in (Hatfield and Hipel, 1999) and in (Dimirovski et al, 2006). Still, the cybernetic structure of negotiation processes does emphasise a comparative framework of knowledge, information, wisdom and uncertainty to be of primary importance. Let for a moment the focus be placed on cases of business negotiation processes, and leave aside more complex policy making negotiations. The information based but qualitative overall structural model is depicted in Figure I. Clearly, other cases of negotiation processes (e.g. socio-political ones) may well have much more complicated models but still follow the same pattern structure as in Figure I.

Apparently, because of its omnipresence in all sorts of decision-making and human social behaviour and let alone in negotiations, the management of uncertainty has become one of the major challenges of our time (Fntowicz and Ravetz, 1993; Klir, 1997; Natke BenHaim, 1997; Dimirovski et al, 2006). In the first place, this is due to decision, control, and monitoring/

72

Buyer

paths usually pass through multiple dilemmas and risks (Dimirovski et al, 2006).

Develop Sourcing Strategy: Identify Sourcing Groups, Evaluate Sourcing Trends

6" '5

I

Supplier

Generate and Distribute RFQs

Analyze RFQs

Analyze Supplier Responses: Scare ft Rank Offers, Identify Optimal Offer Combinations

Formulate Offers

I o

:p

n x: o

Counter Offers, Additional Info

Counter Offers, Additional Info

Negotiation Process

Agree Upon Arrangements

Yes

T Acceptable Deal

No

Reject Noncompetitive Offers

Fig. I. Overall information-flow based model of the negotiation systemic structure and the associated information processing. Furthermore, it should be noted globally the typical behaviour pattern of opponent participants into negotiation processes is structured as shown in Figures II and III; clearly, opponent negotiation participants possessing different skills will have richer structures inside the blocks of thinking, acting, and game engines. For, in practice, decisions during a negotiation process must be made under risk, uncertainty, incomplete, and often even vague information. Thus uncertainly management within negotiation processes has to be kept open to employ fuzzy-set based qualitative information (e.g. see Pete et al, 1998, for instance).

Fig. II. A general two-person game-theoretic model of human behaviour of participants in negotiation processes; solely the thinking sub-process separates us from the animal world. Habitually, scientific societies do have the proper understanding and observe differences among concepts signal, data, information, knowledge, and intelligence following intuition of meaningfulness, measurement evidence, truth, and uncertainty. Yet, negotiation cybernetics involves categories such as fairness, justice, visions of good and evil, and wisdom, and these thinking

Fig. III. Non-cooperative negotiations resemble closely to a combat game model in this illustration based on using finite-state machines with fuzzy attributes. The overall picture of the nature of information that emerges negotiation cybernetics is analogous to the structure and nature of complex systems, as described by complex systems theory. First, it should be noted that the structural model of human behaviour is a feedbackfeedforward one involving both mind and emotions (Mansour, 2001), and this reflects upon social groups and entire societies (Dimirovski, 2001). Hence human centred systems are rather complex systems with a diversity of event driven dynamical phenomena as the main consequence. Secondly, it should be noted that the world is about systemic structure that are inherently non-causal systems, and also only partially identifiable and only partially observable (Mansour, 2001). Should the previous fact on the sole availability of structural models is observed then one may infer (Dimirovski, 2001) they are likely to be only partially controllable too. Furthermore, it should be noted the contemporary mankind's world actually encompasses a network of such interacting systems, typically neither well-defined nor well-structured, constituting the background body of relevance as basis for major conclusions in all areas of socio-economic and socio-politic systems. This is further emphasised by the omnipresence of information category within the framework of negotiation systemic

73

structures, which is a paradigm gaining yet higher importance as the mankind faces the contemporary challenges and risks as demonstrated in the next section.

information, although not both from the same point of departure and not both at the same time. While they established that there exists some overlap in the applicability of the various theories, for the most part each theory appeared oriented towards dealing with a different and specific information deficiency in both theoretical terms and in applications. Their investigations have put a parallel between the nature of information and the nature of complex systems in the sense that both can be placed on scales of complexity, and that both acquire similar attributes as complexity increases. The introduction of systems theory into the field of information management, prompted by the obvious analogies in Shannon information theory, provided a useful guidance to discuss five levels of systems complexity and propose an analogous level of information complexity.

Finally, at this point the basic theoretic techniques of information management hence of uncertainly management too, ought to be recalled. The word is about the theories of probability, fuzzy sets, info-gap models, and Shannon information from the perspective of information management (Natke and Ben-Haim, 1997), of course. Since this investigation study is set within the context of human centred systems, it should be noted, this list of techniques is by no means exhaustive because, in addition to mathematicians and systems engineers, economists, computer scientists and others have 'developed' information-management techniques for their own purposes (Krippendorff, 1986; Haimes, 1993). As a rule, all of these theories do bear a label "uncertainty management technique"; nonetheless, such a label holds true indirectly only. As a matter of fact, each of these theories manipulates data or information, from which the problem solver makes inferences about uncertainly. None of the theories directly addresses the internal perceptions of the decision-maker, i.e. their uncertainly, with regard to the situation at hand, yet these make a core of the negotiation systems.

The simplest complex systems appear to be purely thermodynamic: a system performs in order to remove a thermodynamic gradient from the environment. Such a gradient is essentially a distinction, an indication that one region of the environment can be distinguished from another. To a thermodynamic system, the only information required is the presence and magnitude of this gradient. This was exactly the information being discussed by Shannon information theory: the observable degree of distinguishability. Their next short step was to conclude Shannon information theory is a measure of the simplest type of information.

3. THE NATURE OF INFORMATION: HATFIELDHIPEL THEORY REVISITED Complex systems theory (Kosko, 1993) presents a scale of complexity along which various systems can be placed, from the simplest thermodynamic systems seen in laboratories to intricately complex emergent systems in nature and society. A link between information and systems theory has appeared in the current context as it is possible to note that the simplest kinds of organized systems are completely thermodynamic-like, hence nonseparable fundamental categories of energy and matter along with information, while the simplest kind of information is in the sense of Shannon.

It should be noted, this makes logical sense when considering the complexity increases that have come later in the course of system science developments (Hatfield and Hipel, 1999). Indeed, the first step is to distinguish between simple noise (the environment) and actual information (the gradient). Further, the second evolutionary step in complex systems appeared to be the presence of multiple domains of attraction (or attractors), pertinent to the non-linear nature of real-world processes. Domains refer to the steady states in which the given system can survive, and possessing multiplicity of such domains is the basic indication of the increased complexity. This suggested that the next step in the complexity of information is the concept of relevance implying that certain information is only useful in certain circumstances. The idea emerged from the measure provided by info-gap theory (BenHaim, 1997) of usefulness or efficacy of information for different problems. It may well be noted that a system switching from one domain to another often entails a complete switch in the rules governing the system and its behaviour. This means that the observed system may become dependent on other gradients or other information sources. It would be surprising if this were

In their recent work Hatfield and Hipel (1999) argued that it is the information that should be adopted as the fundamental category, and only then all the facets of uncertainly can be explained while relating the information to truth and meaningfulness. In their exploration, they have found sound observation grounds for the following concluding findings. There are indications affirming the quantity of information available can measured following the information theory (in the sense of Shannon) whereas the novel info-gap models (Ben-Haim, 1997) provide means to measure the quality of the information being examined. Furthermore, traditional probabilistic as well as fuzzy methods are available for modelling the imprecision or error of

74

not so, for the ability to judge the importance or relevance of a piece of data cannot be imparted by method or theory; such know-how is the product of wisdom and is obtained through experience (Kosko, 1993), which is most relevant in the human centred negotiation processes. The third logical step forward into complexity of human centred systems is found when considering the exactness of the data, which is to say vague or precise, indeterminate or not. This was indicated by the very fact that imprecision is set the primary focus of interests in both theories of probability and of fuzzy sets, albeit from different points of departure. For, both theories adopt the distinguishability and relevance in the data being utilized. It should be noted, the sensitivity to precision or spread of data is analogous to a complex system possessing more sophisticated feedback loops and dynamics and being able to modify its behaviour based on the quality of information available (Dimirovski, 2001; Hatfield and Hipel, 1999; Mansour, 2001; Zadeh, 2002). This is precisely the case in the structure of human and society centred systems that are dealt with in social cybernetics. Hatfield and Hipel (1999) rightly noted the essential crucial concepts of information that have been not frequently addressed are those of accuracy, or in qualitative information of correctness. They noticed that all of the existing theories assume the underlying information is inherently true hence all the subsequent findings remain based on that assumption. As a given system grows increasingly sophisticated, however, the question of information correctness must arise. Many human-constructed systems demonstrate awareness of this issue, but also it can be seen in natural systems in the form of genetic reproduction processes. The obvious complexity of any DNA-based organism indicates just how sophisticated a system must be in order to be aware of accuracy issues: indeed, genetic reproduction demonstrates how one can be very precise and very certain yet still be very wrong. This aspect appears rather crucial in a negotiation cybernetic process, which is a human centred system regardless the field and aim of the negotiation (Dimirovski et al, 2006). However, this aspect is already well exemplified in the previous section in the case of business negotiation process (see Figures I and II). How it may appear in socio-political or macro-economic based negotiations it may only involve cybernetic structure of higher complexity. Hatfield and Hipel (1999) have pointed out just as the description of a system is dependent on the perspective equally so is the information observer-dependent, i.e. dependent on the observer who detects and extracts that information, be it

device or human. This also implies that information itself is a subjective concept in general, and let alone in social cybernetic systems and negotiation structures (Dimirovski et al, 2006). The fifth and last step Hatfield and Hipel (1999) have found in questioning the accuracy of one's information, which is a very short yet profound step away from questioning the accuracy of one's analysis of that same information. This commonly leads to the final level of information complexity where information itself begins to re-curse or nest. Just as the most complex systems exhibit emergent patterns of new and unprecedented behaviour, so can questioning existing information lead to infinite levels of new information; this refers to the "recursive" problem explored by Funtowicz and Ravetz (1990) who pointed out how trying to "completely" quantify an uncertainty or imprecision using any tool or approach leads to an infinite series of 'probability distributions of probability distributions' or 'fuzzy sets of fuzzy sets'. This nesting phenomenon appears as much an issue in information management as in complex systems analysis. Hence the only way out is to recognize that the techniques themselves do not provide guidance about when to stop scientific analyzing process. However, as human perception and comprehension gets deeply involved (Zadeh, 2002), the system capacity for indefinite information nesting phenomenon is a tangible danger that may destroy the system. At its best, it can set the system into a cycle of information nesting phenomenon that may launch the system lost which may well happen in negotiation cybernetic social systems (Dimirovski et al, 2006).

4. CONCLUSION The issues of information and uncertainly management within negotiation cybernetic systems have been explored via re-visiting the various concepts related to uncertainty, and by placing them into a comparative framework, including some of the traditional underlying definitions. Such an approach is important not only for the understanding these concepts provide, but also for the purpose to serve as well-defined foundations for further study of decision processes in human-centred systems. Hatfield-Hipel theory on information and uncertainly management was re-visited in a way as to place it within the dynamic phenomenology of centred negotiation processes in human society centred systems. From a social cybernetic perspective, both certainty and uncertainty are not measurable properties of data, but rather they are a part of the decision-makers' perceptions hence prone to subjective handling. Hence it is the composite fuzzy-linguistic and probabilistic view on

75

Howrad, N. (1971). Paradoxes of Rationality: Theory of Metagames and Political Behaviour. MIT Press, Cambridge, MA. Kile, F. (2005). "Investigation into control of the global social system." In: Preprints of the 16th World Congress of the IFAC (P. Horacek, M. Simandl, P. Zitek (Eds)), Prague, Fr-M21-TO/3.(l-6). IFAC and Czech Technical University, Prague, CZ. Klir, G. J. (1997). "Uncertainty theories, measures, and principles: an overview of personal views and contributions." In: Uncertainty: Models and Measures. (H. G. Natke and Y. Ben-Haim (Eds)), pp. 27-43. Akademie-Verlag, Berlin, DE. Kopacek, P., and L. Stapleton, Editors (2004). Technology and International Stability - Proceedings of SWIIS 2003. Elsevier Science, Oxford, UK. Kosko, B. (1993). Fuzzy Thinking. Hyperion, New York. Krippendorff, K. (1986). Information Theory: Structural Models for Qualitative Data. Sage University Paper on Quantitative Applications in the Social Sciences, Series no. 07-062. Sage Publications, Beverly Hills, CA. Kreps, D. V. (1995). Game Theory and Economic Modelling. Clarendon Press, London, UK. MacFarlane, A. G. (1993). Information, Knowledge, and Control. In: Essays in Control - Perspsctives in the Theory and Its Applications (H. L. Trentelmann and J. C Willems (Eds)), pp. 1-28. Birkhauser, Boston, MA. Mansour, M. (2001). "Systems theory and human science" (Plenary paper). In: Conflict Management and Resolution in Regions of Long Confronted Nations (G. M. Dimirovski (Ed)), pp. 1-10. Elsevier Science, Oxford,UK. Natke, H. G., and Y. Ben-Haim (1997). Uncertainty: Models and Measures (H. G. Natke and Y. Ben-Haim (Eds)). Akademie-Verlag, Berlin, DE. Nash, J. F. (1951), "Non-cooperative games." Ann. of Mathematics, 54, 286-259. Osborne, M. J., and A. Rubinstein (1994). A Course in Game Theory. MIT Press, Cambridge, MA. Pete, A., K. R. Pattipati, D. L. Kleinman and Y. N. Levchuk (1998). An overview of decision networks and organizations. IEEE Trans, on Syst., Man & CyberneticsPt.C: Applications & Reviews, 28 (2), 173-193. Petit, M. L. (1990). Control Theory and Dynamic Games in Economic Policy Analysis. Cambridge Univ. Press, Cambridge, UK. Royal Society (1992). Risk: Analysis, Perception, and Management. Report of a Royal Society Study Group, The Royal Society, London, UK. Russell, B. (1975). Power. Unwin Books, London, UK. Wagner, R. H. (1994). "Peace, war, and the balance of power." American Political Science Review, 88 (3), 593-607. Zadeh, L. A. (1973), "Outline of a new approach to the analysis of complex systems and decision processes." IEEE Trans. Systems, Man & Cybernetics, 3 (1), 28-43. Zadeh, L. A. (2002). "Toward a perception based theory of probabilistic reasoning with imprecise probabilities." J. of Statistical Planning & Inference, 105, 233-264. Zimmermann, H.-J. (1997). "Uncertainty modelling and fuzzy sets". In: Uncertainty: Models and Measures (H. G. Natke and Y. Ben-Haim (Eds)), pp. 84-100. AkademieVerlag, Berlin, DE.

information that is the crucial category in social systems cybernetics, in general, and in negotiation systems, in particular. REFERENCES Ayres, R. U. (1994). Information, Entropy, and Progress: A New Evolutionary Paradigm. American Institute of Physics, New York, US. Basar, T., and G. J. Oslder (1999). Dynamic Non-cooperative Game Theory (2nd ed). The SIAM, Philadelphia, PA. Ben-Haim, Y. (1997). "Beyond maximum entropy: exploring the structure of uncertainty." In: Uncertainty: Models and Measures (H. G. Natke and Y. Ben-Haim (Eds)), pp. 1126. Akademie-Verlag, Berlin, DE. Brams, S. J. (1994). Theory of Moves. Cambridge University Press, Cambridge, UK. Dimirovski, G. M., N. E. Gough, and S. Barnett (1977). "Categories in system and control theory." Intl. J. Sys. Science, 8(9), 1081-1090. Dimirovski, G. M. (2001). "Applied system analysis to transition crisis of Southeast Europe and new national goals formation". In: Conflict Management and Resolution in Regions of Long Confronted Nations (G. M. Dimirovski (Ed)), pp. 85-93. Pergamon Elsevier, Oxford, UK. Dimirovski, G. M., A. T. Dinibiitun, F. Kile, R. Neck, J. Stahre, and L. B. Vlacic (2006). "Control system approaches for sustainable development and instability management in the globalization age." Annual Reviews in Control, 30 (to appear). Dinibutiin, A. T., and G. M. Dimirovski (2001). "Technology transfer to developing countries and technological development for social stability - Part 1." In: Proceedings the 8th IF AC Workshop on Supplemental Ways for Improving International Stability (P. Kopacek and M. W. Han (Eds)), pp. 25-26. Elsevier Science, Oxford, UK. Eeckhout, J. (2000). Competing Norms of Co-operation. Department of Economics, University of Pennsylvania, Philadelphia, PA. Funtowicz, S. O., and J. R. Ravetz (1990). Uncertainty and Quality in Science for Policy. Kluwer Academic Publishers, Dordrecht, NL. Funtowicz, S. O., and J. R. Ravetz (1993). "Science for the post-normal age." Futures, 25 (7), 739-755. Gibson, J. L., J. M. Ivancevich and J. H. Donnelly Jr (1997). Organizations: Behaviour, Structure, Processes. Irwin, Homewood, IL - Boston, MA. Gitt, W. (1987). "Information: the third fundamental quantity." Siemens Review, 56, 36-41. Glenn, C. and M. Hannan (2000). Demography of Corporations and Industries. Princeton Univ. Press, Princeton, NJ. Haimes, Y. Y. (1998). Risk Modelling, Assessment, and Management. J. Wiley, New York, US. Hatfield, A. J., and K. W. Hipel (1999). "Understanding and managing uncertainty and information," In: Proceedings of the 1999 IEEE Conf. on Systems, Man, and Cybernetics, 12-15 October, Tokyo, vol. 2, pp. V-1007 V-1012. The IEEE and Tokyo Institute of Technology, Tokyo, JP.

76

Copyright © Improving Stability in Developing Nations through Automation Prishtina, Kosovo, 2006

ELSEVIER

IF AC PUBLICATIONS

THE ECONOMICS AND POLITICS OF ENERGY GENERATION

M.A. Hersh Electronics and Electrical Engineering, University of Glasgow, Glasgow G12 8LT, Scotland. Tel: +44 141 330 4906. Fax: +44 141 330 6004. Email: [email protected]

Abstract: This paper uses the example of the UK to discuss the political and economic context in which energy is provided and the effects of this context on energy policy. It evaluates the advantages and disadvantages of renewable energy sources and their role in a sustainable and socially just energy policy. It shows that the relative economics of different energy sources depends on the how their costs are calculated, with discounting and market forces posing a barrier to renewables. Copyright © IF AC 2006. Keywords: Energy, economic, political context, renewables

1.

INTRODUCTION: ENERGY SUPPLY AND DEMAND

•

There is increasing concern about energy supplies, arising from a mixture of factors, including political instability and political changes which may affect the availability of oil and concerns about global climate change associated with anthropogenic (human generated) emissions of carbon dioxide and diminishing supplies of fossil fuels (Hersh, 2006). Consequently there is a need for appropriate alternative strategies to ensure energy supply. This paper will present a case study of energy issues in the UK as a mechanism for introducing discussion of the wider issues.

Heat, to only a limited extent, distributed as hot water or steam. There are a few district or community heating schemes, which provide hot water or steam to buildings from centralised boilers. A small amount of heat is obtained from geothermal sources.

Machinery and information technology are most conveniently powered by electricity, whereas applications involving heating should use other forms of energy than electricity to avoid the energy losses associated with conversion and transmission. The electricity supply needs to meet the following two specifications: • Sufficient capacity to meet overall energy requirements • Electricity provision when and where it is needed.

In the UK energy is delivered in the following five main forms (Boyle, 2004): • Liquid fuels, largely oil and its derivatives (petroleum, diesel, kerosene). • Gaseous fuels, mainly methane (natural gas), plus bottled propane and butane. • Solid fuels, mainly coal and its derivatives, with small amounts of fuel wood, refuse incineration and straw burning. • Electricity, mainly from fossil fuels and nuclear power stations. Hydroelectricity and other renewable sources make a small contribution.

There is a need both to reduce overall energy demand and to manage it to reduce the variation in demand over the base load. There are a number of approaches to reducing energy consumption, including the following: • Energy efficient buildings, appliances and lighting which are designed to minimise energy use. This will probably require financial support to upgrade existing buildings and stricter

77

•

•

•

•

building regulations for new buildings, as well as financial support or loans for the purchase of energy efficient appliances which save money overall, but may have higher capital costs. Energy efficient heating systems, as well as heating buildings to slightly lower temperatures. Financial support may be required for upgrading existing heating systems and there may also be a need for new regulations and/or legislation on energy and emissions specifications. Heating buildings to slightly lower temperatures will require people to wear more clothes and to be more active and may be inappropriate for some elderly people or young children. The use of adaptive and other advanced control schemes in industry to reduce energy and material consumption (which has associated energy costs). Reduced consumption, improved waste management and design for waste minimisation, refurbishment and repair, reuse and recycling. The resulting reduction in waste, resource use and the differences between the energy requirements of refurbishment and repair and manufacture of new products will result in energy and carbon dioxide emissions savings. A reduction in transport energy consumption. This will require measures to promote a modal shift from private car use to cycling, walking (or wheelchair use) and public transport, an encouragement of facility accessibility rather than mobility and reductions in air travel. Moves to accessibility rather than mobility would require an end to the resiting of facilities, such as shopping and sports centres, away from the population centres they serve and moves to resite out-of-town facilities close to centres of population, though this would have to be implemented over a period of time. Reductions in air travel will require rail travel to become more competitive with air travel (including by changes in their relative prices, infrastructure development to allow the widespread use of very high speed trains and further technical developments to increase the speed and reduce the environmental impacts of train travel), an increased use of video conferencing to replace business travel and possibly the introduction of tradable air travel permits.

behaviour, for instance an increase in demand of over 2 GW in a few minutes as large numbers of people switch on the electric kettle at the end of a popular television programme, as television power consumption is relatively low compared to that of an electric kettle (Boyle, 2004). Existing large coal and nuclear power plants do not respond well to a fluctuating demand, as they may take 24-36 hours to reach their full output from cold. A changing demand could be followed by running some stations at part load, but this would reduce their overall efficiency and increase carbon dioxide emissions per kilowatt of electricity generated. A smaller combined cycle gas turbine station can produce some power within an hour, but will take about 8 hours to reach full output. Therefore pumped storage stations are generally used to respond to variations in power demand. There are currently three pumped storage plants in the UK, two in Wales and one in Scotland. The combined peak output is over 2 GW or about 5% of the typical winter electricity demand, but they are only designed to operate for short periods until conventional plant is brought on line (Boyle, 2004).

2. RENEWABLE ENERGY Renewable energy can be defined as energy flows which are continuously replenished by natural processes. This is in distinction to fossil and nuclear fuels and oil, which consist of energy stocks, of which only finite quantities are available, rather than flows. Renewable energy contributed about 10.9% of global energy requirements in 2002 and this figure is growing at 1.5% per annum (World Bank, 2004). Renewable energy sources can be divided into the following three main categories: • Direct uses of solar energy i.e. solar thermal energy conversion and solar photovoltaics • Indirect uses of solar energy i.e. hydropower, wind power, wave power and bioenergy or biofuels. • Sources of renewable energy which do not depend on solar radiation i.e. tidal and geothermal energy. Since they are energy stocks, fossil and nuclear fuels and oil will be exhausted at some point. The quantity of fuel that can be extracted and therefore the length of time until exhaustion depend on the availability of technology that can extract the fuel and do so at a price that is considered economically acceptable. Advances in technology will increase the amount of fuel that can be extracted and improve the economics of this process. However attention should be paid to the environmental and social impacts of this technology and the extraction process. In addition increasing scarcity may change the economics of fuel extraction and significantly increase the price at which fuel extraction is

Demand management should include measures to encourage the transfer of energy consuming activities to times of low demand to reduce the variation in total demand. However, it will probably still be necessary to have some additional capacity to cope with demand variations. Currently demand varies both seasonally and over the day, with an average load of slightly less than 40 GW, which can drop to under 20 GW on a summer night and rise to over 50 GW on a cold winter evening. This gives a base load of about 20 GW plus a varying additional load. Localised peaks also occur due to synchronised

78

considered economically viable. However, the extent to which technology can increase the availability of fossil and nuclear fuels and oil is limited and therefore a technological 'fix' is not the answer.

of green house gases, mainly carbon dioxide and the acid rain gases, sulphur dioxide and nitrogen oxides than fossil fuels and have no or minimal emissions of these gases when operating to generate heat or electricity. There is now a consensus of scientific opinion, as expressed by the International Panel of Climate Change (Houghton et al., 2001) that anthropogenic (human generated) emissions of greenhouse gases are causing global warming and will lead to an average temperature rise of 1.4 to 5.8 °C by the end of the century. The consequences of global warning will include rises in sea levels and the drowning of low lying areas, more frequent climatic extremes, the disruption of agriculture and ecosystems and possible insecurity of food supplies (Hersh, 2006; McCarthy et al., 2001). The use of renewable energy sources has already reduced emissions by 1,500 million tons of carbon dioxide each year or 7% of current energy related emissions. Between 6000 and 9000 Mt of emissions (corresponding to 40% of energy related emissions) could be avoided by 2020 (OECD, 1998). Policy measures to reduce or at least stabilise energy consumption in combination with a staged increase in the percentage of energy generated by renewable sources to 100% would lead to further significant reductions in the emissions of greenhouse gases.

Since renewable energy sources are flows, a given amount is available each and every year and therefore it is not possible to overconsume, as the amount of energy extracted from the flow in a given period does not affect the volume of flow in subsequent periods. For instance using a flow of water down a gradient to generate electricity in a given period will not reduce the flow of water and its potential energy generating capacity in subsequent periods, though this may be reduced by climatic changes such as reduced rainfall or increased temperatures leading to a (partial) drying up of the watercourse. The one exception, where renewability depends on renewable rates of consumption is geothermal energy. Although geothermal resources are a natural energy flow rather than an energy store, they can be and frequently are mined in a nonrenewable way. Heat is generally extracted faster than it can be replaced, leading to declining temperatures in producing steam fields. This then reduces the temperature at which water and other fluids are stored in interconnected rock cavities and constantly the quantity of heat energy that can be extracted.

Renewable energy sources have the further benefits of not generating radiation, including radioactive waste products, and avoiding danger to workers in the mining and milling stages. The production of nuclear power produces radiation and radioactive waste at all stages of the life cycle. The problem of the safe disposal of radioactive waste has not yet been solved and seems unlikely to be in the foreseeable future. Coal mining also produces a limited amount of radiation. The dangers to workers associated with coal mining are well known, but the fact that uranium mining and milling is the most hazardous life cycle stage of nuclear power production is generally ignored. The risks to workers and their families include cancers of various types, sterility, stillbirth and genetic damage to children. At the time of highest mine activity the lifetime risk of fatal cancer to workers was sometimes as high as one in nine (Dropkin et al., 1992).

At least in theory, renewable energy sources provide potentially unlimited sources of energy. In practice, there are technical and economic limits on the energy that can be currently extracted. However, these limits are the result of a particular political context in which research funding has been allocated preferentially to nuclear power and the way in which energy sources are costed actively disadvantages renewable energy, as will be discussed in more detail in Section 3. Currently the richest fifth of the world's population is responsible for 58% of global energy consumption, whereas the poorest fifth only uses 4% of global energy (Michaelis, 2002). The limited supplies and distribution patterns of fossil fuels mean that the current heavy energy consumption by the richer countries reduces its availability to the poorer countries. This is not the case with renewable energy sources, due to the (potentially) unlimited availability of renewable energy and therefore renewable energy sources have the potential to remove the current inequalities in access to energy. Currently over half of energy in sub-Saharan Africa and nearly two fifths in South Asia is generated from renewables and waste (World Bank, 2004), further showing the potential of these sources in the low income countries. Renewables could also accelerate the electrification or energisation of rural communities in these countries.

The degree of risk will depend on the measures taken to protect workers and their living conditions, which will largely be determined by the social, political and economic context in which uranium mining and milling take place. For instance there were considerable differences in the housing provided for black and white workers at the open cast Rossing uranium mine in Namibia under South African occupation (Moody, 1992). The fact that there has been less interest in and less information published about health, safety and environmental issues at the Rossing mine in the period since Namibian independence illustrates another aspect of the

Renewable energy sources have a number of other benefits in addition to being inexhaustible. In particular they have much lower life cycle emissions

79

political context in which the energy debate takes place. The numbers of injuries per year has reduced by a factor of 6 between 1988 and 2004 (Rossing, 2004), indicating that the changed political context seems to have had a major impact on health and safety.

sensitivity in site selection, use of the best available technology and including the public and appropriate organisations in the early planning stages, need to be given serious consideration in decisions about the most appropriate renewable energy systems for a given area.

Further socioeconomic and environmental benefits of renewables include the following (OECD, 1998): • Reducing dependence on fuel imports and diversifying and securing indigenous energy supplies. • Promoting the decentralisation of energy by providing small, modular, rapidly deployable schemes. • Providing job opportunities in rural areas. • A reduced need for upgrading electricity distribution systems or building new line capacity due to the modular and distributed nature of renewable energy systems, resulting in a reduction in construction and transmission emissions and energy consumption or losses and reduced visual intrusion.

3. THE ECONOMICS OF RENEWABLES The viability of any energy generation scheme depends on the following factors: 1. Technical factors 2. Social and environmental impacts 3. Economic factors In the past the main focus was on technical and economic factors. However there is an increasing recognition of the importance of environmental factors and, in particular, the importance of avoiding emissions of greenhouse gases and long term security of supply when fossil fuels run out. This is leading to increasing interest in renewable energy. Energy supply takes places in a political and economic context which may promote particular sources of energy and disadvantage others. It should be noted that there may be discrepancies between avowed policy and practical measures and that there may be political or economic vested interests in promoting a particular type of energy generation. For instance, the UK government has come out in favour of building more nuclear power stations, whereas their independent advisory body on sustainable development, the Sustainable Development Commission (SDC) has concluded that nuclear power will not solve the problem of global warming and only lead to limited reductions in carbon dioxide emissions. The SDC is also concerned about the lack of long term solutions for the disposal of nuclear waste, the uncertain economics of nuclear power, the UK being locked into a centralised distribution system for 50 years, energy efficiency being undermined and issues of international security (SDC, 2006). For many years several times as much research funding has been available for nuclear power as for renewable energy (Ahern, 2001). This lack of funding has meant that renewable energy technology is not as advanced as it could have been. In effect the disparity in funding has promoted nuclear power at the expense of renewables.

There are also environmental benefits associated with specific renewable energy systems. For instance, hydroelectric schemes can improve water supplies and over one million small wind turbines already pump water from underground reservoirs. Growing energy crops can reduce soil erosion and, when grown on land degraded by inappropriate agricultural practices, improve soil conditions and enhance wild life diversity. Some renewable energy technologies can reduce urban pollution by providing power for electric vehicles, though this should not replace moves to reduce transportation energy, particularly the high energy consumption associated with private car use. Several renewable energy technologies could be used to produce water by desalination. Although, as discussed above, most renewable energy systems have numerous environmental and socioeconomic benefits, they also have a range of negative environmental impacts, many of which are small, localised and site specific (OECD, 1998). If appropriately designed and sited and using renewables to provide any energy inputs, many renewable energy systems have low or very low environmental impacts in the operation life cycle stage. However there may be significant environmental impacts in the construction life cycle stage, with these impacts similar to those of other engineering projects of a similar scale. Exceptions are large scale hydro and tidal schemes, which can also have very significant negative environmental impacts in operation. The construction and, for some renewable energy systems, also the operation phase involve inputs of energy, which should preferably be generated by renewable sources. The likely environmental impacts, as well as the likely effectiveness of mitigation measures, such as

There has been considerable pressure in the European Union to treat electricity as a totally free market commodity. Prior to 1989 the UK electricity system was under national ownership and had three main components (Boyle, 2004): • The Central Electricity Generating Board (CEGB) power stations which supplied the electricity. • The National Grid, also part of the CEGB, which distributed electricity at high voltage.

80

•

required to promote a combination of energy efficiency and conservation with a supply based totally on renewables, as the only way to achieve long term security of supply in an environmentally and socially sustainable way that does not contribute to global warming or inequalities in access to energy. (The potentially unlimited availability of renewables was discussed in Section 3.) This policy should involve supply diversity in terms of a mix of different types of renewable energy sources located in a variety of different areas. This would have the advantage of smoothing out variations in supply and improving supply security, as problems with or threats to one type of supply are unlikely to affect others. The costs of more expense renewable resources could be balanced against those of cheaper ones to give an acceptable average cost.

The local Area Boards which distributed low voltage electricity to customers.

In 1989 the UK electricity industry was privatised and these parts were split up and sold to a large number of private companies. This did not include the older nuclear power stations, which were recognised not to be economically viable. Since privatisation a number of mergers and take-overs have taken place and some companies have invested in the US and/or have been taken over by overseas companies. For instance the state-owned Electricite de France bought London Electricity. Currently decisions about which stations run at a given time are made by computer and based on a continuous process of competitive bidding. A computer model run by the private National Grid Company (which took over the national grid) estimates demand for the next few hours and invites bids to supply electricity. The different power station owners supply bids through computer programs, the cheapest offers are accepted and the appropriate stations are brought on line by remote control. Scotland and Northern Ireland were originally outside this competitive bidding pool, but the British Electricity Trading and Transmission Arrangements (BETTA) brought Scotland into the pool in April 2005.

Only governments are in a position to develop and implement this type of strategy. However they will require pressure from concerned individuals and organisations to do so. Market mechanisms will only promote the energy technologies that are most profitable under existing pricing structures. Government subsidies, legislation and regulation can have a limited influence on pricing and be used to require private firms to supply a certain proportion of renewables, but not to do so in a coordinated way or to the extent required. Market economics based on discounting of the future value, social, environmental and other costs of energy often act as a disincentive to the introduction of renewables. Therefore the introduction of a coordinated energy policy to increase the proportion of energy generated by renewables and promote energy efficiency, as well as a pricing structure to implement social and environmental objectives will only be possible under pubic ownership. This type of pricing structure could be used to encourage low energy consumption, while penalising high energy consumption, and could even provide a certain amount of energy free of charge to some or all domestic consumers. This would avoid the current tragic and avoidable situation of a number of elderly people dying of hyperthermia each winter.

Like many other countries, the UK has financial incentives for renewables, including the following: • Capital grants, which are available on a competitive basis for planting energy crops and offshore wind, solar thermal and PV systems. • The Renewables Obligation, which replaced the Non-Fossil Fuel Obligation (NFFO) in 2002. It puts a statutory duty on electricity supply companies to obtain a certain proportion of their electricity, which will be increased to 10% by 2010, from renewable sources. Additional costs of up to 3p/kWh can be passed on to users, but are unlikely to add more than 4% to bills. • Exemption from energy taxes. • Renewable Energy Feed-In Tariffs (REFIT) fixed premium prices for electricity from renewable sources. • The Climate Change Levy (CCL), which was introduced in 2001 for energy supplies to business, but not domestic consumers. New renewable sources and 'high quality' combined heat and power systems, but not nuclear power, are exempt.

The direct costs of any energy generating system are based on the following four components: the capital costs of constructing the system (including any interest on loans), fuel costs, operation and maintenance costs and decommissioning costs. For most renewable energy systems, the fuel costs are zero and the capital costs are relatively high and the most significant cost element. Many renewable energy systems have long lifespans. Unfortunately discounting, based on the idea that a sum of money today is more valuable than the same amount in the future, is used in financial calculations and often applied to the value of the electricity generated, reducing it to zero after a number of years which depends on the discount rate used.

However, despite these incentives, the basic context of energy policy and infrastructure development in the UK is competitive bidding by a number of private companies. This context does not support the development and implementation of a coordinated energy policy based on long term social and environmental needs rather than short term economic advantage and including an appropriate energy infrastructure. This type of energy policy will be

81

Most renewable energy systems have high capital costs, relatively low running costs, no fuel costs and long life times. For schemes with long life times, such as over 100 years for hydroelectric plants, the capital repayment period (rarely more than 25 years) is much shorter than the plant life time. Therefore, if the running costs were ignored, the plant could be considered to provide free electricity after 25 years, but the value of this electricity is often discounted to zero. Therefore the existing market approach to evaluating the costs of renewable energy systems generally disadvantages them due to the high initial costs and the use of discounting. This means that the relative economics of renewable and fossil fuel or nuclear systems would change under a different approach to costing.

systems. Diversity of supply has a number of benefits including smoothing variations in supply, making it easier to follow variations in demand and increasing security of supply. Accounting procedures which take account of the full environmental costs of fossil fuels and nuclear power, including the future costs of decommissioning nuclear power stations and storing the radioactive waste for very extended periods. Significant financial support, possibly in the form of low or zero interest rate loans, for the construction of renewable energy systems.

Therefore the economics of a particular type of energy depend on the following factors: • The costs: capital, fuel, operation and maintenance, and decommissioning. • Whether energy generation is privatised or nationalised. Private firms generally have to make a profit in the short term (while operating within the law), whereas nationalised systems are able to take into account environmental and social issues, such as the imperative to reduce carbon dioxide emissions. • Whether discounting is applied to the value of the electricity generated and the discount rate used. • How the environmental and social costs of generation are taken into account, as this will affect the relative costs and economics of the different systems. • The availability of subsidies, tax discounts and other financial or policy support for renewable energy. • Whether the source of energy is required to be competitive with other sources of energy, make a profit or cover costs.

This paper has discussed the economic and political context in which decisions on energy policy are taken in the context of a UK case study. It has demonstrated that this context has a determinative effect on the resulting energy system. In particular it has been shown that a sustainable energy policy based on renewables, energy efficiency and energy conservation will only be possible under national ownership and that current techniques for calculating the costs of energy, including the use of discounting, act as a barrier to renewables.

4. CONCLUSIONS

REFERENCES Ahern, N. (2003). The European Union's support for renewable energy. In: R. Douthwaite (ed.), Before the Wells Run Dry, Green Books. Boyle, G. (ed.) (2004). Renewable Energy: Power for a Sustainable Future, 2 nd ed., Oxford University Press and Open University. Dropkin, G. and D. Clark (1992). Past Exposure: Revealing health and Environmental Risks of Rossing Uranium, Namibia Support Committee and Partizans, London. Hersh, M.A. (2006). Mathematical Modelling for Sustainable Development, Springer Verlag. Houghton, J.T. et al. (eds.) (2001). Climate Change 2001: Scientific Basis, Cambridge Univ. Press. McCarthy, J.J. et al. (eds.) (2001). Climate Change 2001: Impacts, Adaptation and Vulnerability, Cambridge University Press. Michaelis, L. (2002). Sustainable consumption and production. In: Dodds, F. (ed) Earth Summit 2002 a New Deal, Earthscan, 75-100. Moody R. (1992). The Gulliver File. Minewatch. OECD (1998). Benign Energy? The Environmental Implications of Renewables, OECD/IEA. Rossing (2004). Rossing Working for Namibia Report to Stakeholders, www.rossing.com. SDC (2006). The role of nuclear power in a low carbon economy, Sustainable Development Commission, http://www.sd-commission.org.uk. World Bank (2004). World Development Indicators, The World Bank.

For renewable energy sources to provide a majority and eventually almost all the world's energy requirements a change in the relative economics of renewable and other energy sources will be required. This will involve one or more of the following: • A move away from current approaches based on discounting, so that the full life time value of the energy generated can be considered. This will probably require nationalisation of energy provision, as the main concern of most commercial firms is relatively short term profitability. • Balancing the costs of more expensive renewable energy sources against those of cheaper ones, so that the average cost of the power generated is acceptable, though the costs of some units may be high. This would again require either nationalisation or significant subsidies or guaranteed prices for the more expensive energy

82

Copyright © Improving Stability in Developing Nations through Automation Prishtina, Kosovo, 2006

ELSEVIER

IF AC PUBLICATIONS

ECONOMIC, ETHICAL AND CULTURAL ASPECTS OF ICT CHANGE Jozef B. Lewoc1, Antoni Izworski2 and Slawomir Skowronski2 Leader (Leading Designer), ul. Powst. SI. 193/28, 53-138 Wroclaw, Poland Wroclaw University of Technology, Wybrzeze Wyspianskiego 27, 50-370 Wroclaw, Poland Corresponding author: Jozef B. Lewoc, ul. Powst. SI. 193/28, 53-138 Wroclaw, Poland Tel: +48-71 361 4436, Fax: +48-71- 783 7897, e-mail: [email protected]

Abstract: The economic, ethical and cultural aspects are most clearly seen during the technology transfer and change processes. The paper presents a case study of such processes, developed in a high degree a case study country of Poland, after the political transformation in 1989. The paper describes several pitfalls that may be fallen in by both the technology providers and recipients. The pitfalls are illustrated by actual events occurring during the ICT technology transfer in power systems in the case study country. Copyright © 2006 IF AC. Keywords: electric power systems, energy distribution, computer control system design, technology transfer, methodology

a dozen or so percent. Assuming ca 10% benefits, we obtain the value of the benefits equal to some 40,000,000,000 dollars.

1. INTRODUCTION The problems of ICT implementation in automation, raised by the SWISS Conference Prishtina 2006 are both economically and socially important. On one hand,, understanding of them may help for the systems designers to think over what they actually work for, what can be the impacts of their work and what to do that their work is accepted and included. On the other hand, understanding the problems may help many ICT recipients may defend themselves against inappropriate methods employed sometimes by big corporations delivering ICT systems.

Therefore, the paper is a case study of the economic, ethical and cultural effects of the ICT transfer process in the power system automation in the case study country. The economic, cultural and ethical; aspects of technology change and transfer are discussed by many authors, especially of the SWIIS (Supplemental Ways for Improving International Stability) circles (Hersh, 2001a; Jancev, 2001). The authors of the present submission believe it to be appropriate to support these generalisations with some detailed data acquired during the on-going process of the technological change for ICT in computer automation for the power industry in the case study country.

Applications of ICT in computerised automation systems in the power industry is a very important domain from the social point of view. This is implied by very big economic and financial effects of implementation of ICT in the domain. For instance, in the case study country of Poland, regional power distribution board computer monitors (Monitors) implemented in the late seventies had under their control more than a half of electrical power in the country for close to 20 years (Lewoc, 1992). In the power industry, economical benefits are evaluated as

The intention of the authors is to blow the whistle (Hersh, 2001b), i.e. to warn of the pitfalls that may be found on the way towards the ICT change that, at least in theory, should lead the society to some

83

magnificent future. In other words, the authors do believe that warning of such pitfalls may help in realisation of the expectations (and, perhaps, dreams) of the future social stability (Kile, 2001).

hard conditions then their counterparts in the Western countries. This approach was forced by the general situation of the case study country but was somewhat similar to that in the University of Cambridge Cavendish Laboratory, where the lowcost approach to science was their tradition. For instance, Sir Marin Ryle (diseased) was proud to say about the minimum financing for their work. So they could have a chance to develop themselves well: more then half of sky objects were identified by C for Cambridge and they got the Nobel prize for pulsars and super-synthesis of aperture. And the case study country got the Monitor benefits. Without a serious approach to performance evaluation, it would not be possible

The technological change process under discussion began, definitely, in Poland in the seventies though some work was done earlier, trying to develop a power generating unit monitor on some obsolete computer with drums as the basic operating memory. The task was, obviously, infeasible.

1.1 Pitfall 1 If you develop a novel computer system (novel means here a pioneering system at least in your country) and you forget the slogan: "Think computers, think performance ", your efforts may be useless.

1.2 Pitfall 2 Jumping into a conclusion that an ICT large-scale system, even the one manufactured by some world leader in the area, must be better that the system developed in your country, directly for your fellowcitizens, results often in high losses.

Poor performance characteristics were the reason of rejection (refused acceptance) of many ICT implementation projects, especially in power system automation. A good example may be numerous problems with application of PC compatibles in the case study country and elsewhere. However, the computer, Odra 1325, available in the seventies in the case study country (ICL 1902a compatible), though intended primarily for numerical computations and of much less computing power, enabled to initiate, actually, the very successful technological change. But this required rather a lot of work. First of all, a system of industrial peripherals (SMA) was to be developed with a special consideration paid to adequate field data transfer throughput. Secondly, a special-purpose operating system was designed, also considering performance an significant characteristic of the system. The whole configuration (hardware and software) was investigated by simulation and ad hoc measurements to verify that the performance constraints are not violated.

The early eighties were the years when the technological change was stopped in a considerable degree due to political reasons (the marshal law). However, the local staff educated in the seventies developed successful systems (training simulators, microprocessor-based monitoring tools for power distribution and generation centres, etc.). The training simulators, developed on the on the same hardware and software solutions, though, of course, adopted to the needs of e-training, deserve some special notice. Though it is commonly assumed that effective training simulators need computing power of several MIPS, the systems developed in the case study country on the Odra 1325 computer that was some three orders of magnitude slower, proved very useful (more than 90% of Polish power network engineers were successfully trained on the simulators). This was possible due to the fact that instead of solving complicated differential equations, pre-prepared answers defined by experienced trainers were used in e-training sessions and the problem of inadequate computing power was successfully by-passed.

It should be mentioned here about some hardware and software tricks that must have de devised by the designers to overcome the problems created by the rather obsolete hardware and software they had at their disposal. For instance, poorly designed core memory drivers in Odra 1325 resulted in frequent trip-offs of the computer in operation. Considering such problems, the design team developed the software in a way enabling high-speed restart and a simple restart PC board plus a piece of software enabled to by-pass the problem effectively: the overall availability factor of the systems in regional power distribution board exceeded 90% during some eighteen years (maintenance breaks included).

Another case exemplifying this pitfall may be that of o computer integrated manufacturing and management (CIMM) system for a new power plant, that was designed in late seventies. However, the power plant constructions was delayed severely. Some five years later the contract for of automation of the power plant was granted to a big Western corporation. However, they could offer the low level monitoring system only, roughly a dozen or so percent of the complete CIMM hardware and software architecture.

When the problems connected with Odra 1325 reliability are added, one can easily see that the designers of Monitors were to work in much more

84

Development of the "managerial" part of the system later and attempts to integrate it with the monitoring part of the system at a later date, proved to be rather costly and hardly feasible.

In a part, it was due to a down economy period. However, as in the case of the institute, that had been earlier in a convenient position of the monopolist in development of ICT systems in the power industry, a major reason was that the institute could not and even did not try to fight for its leading role in implementation of ICT systems in power system automation.

As a matter of fact, the offer of CIMM systems, severely needed by the power industry and other enterprises is still empty even now (2006).

The institute suffered a lot: they employed some 700 hundred people in the most successful period of their activity. Now they employ a hundred or so people and perform rather minor tasks for the power industry.

1.3 Pitfall 3 If you do not use your best people to solve difficult design and implementation tasks, you may suffer severe losses. In the late eighties, some new work on large-scale technological changes was done with various results. The leading positive role was played there by a local institute involved in power system control that had successfully designed and implemented the ICT systems mentioned above. The institute possessed a staff of the most successful experience in design and implementation of ICTs in power system automation in the case study country. The experience included a power generating unit monitor for a power plant.

1.5 Pitfall 5 When you try to transfer ICT systems and do not make use the best local staff, you may lose. Considering the negative effects connected with the ICT change in the power system automation in the case study country, mentioned above, people involved in the processes, especially the leading designers who had been the actual parents of the successes, waited hopefully for possible big corporation to enter the case study country to carry out the ICT transfer job. The assumption was that they had been successful with similar their own systems so they should be useful for the new ICT system transfer work. This assumption was reasonable since the leading designers were the ones who knew the best the actual problems connected with acceptance of ICT systems in the case study country, and the system acceptance is ranked by many experts as the most important feature of any ICT system.

However, when an order was received to design and implement a monitor, on a new minicomputer, for the biggest power plant in the country, the task was allocated to a team of no actual earlier experience in the problems. The results were pitiful. The system development period was about 8 years. During the period, the leading designers changed several times. The design documentations looked as a collection of mutually independent and often inconsistent parts. And the work was completely unsuccessful. In addition to the loss of money, the institute lost a lot of their prestige as the leading institution in design and development of ICT systems for the power industry in the case study country.

Unfortunately, that was not the actual case. In the nineties, four big corporations operating in the power system automation established their branches in the Lower Silesia (Wroclaw province. However they were not interested in employing the leading designers. Instead, they were interested in buildings and equipment, managerial staff (tested with a negative result) and lay labour.

Another example may be here that of the power system computer automation centre established in Wroclaw wherein the institute had its seat. Some staff was taken from the institute but from the department possessing no earlier actual experience in design of ICT implementations in power system automation. The effect was that for many years they were only capable to develop cabling for ICT systems developed by other companies.

Corporation 1 took over the Polish biggest manufacturer of computers, declaring to help them to remain in the computer market. The corporation never tried to find the successful leading designers of ICTs, the actual pioneers of the computer science in the case study country. Consequently, the corporation was not successful to find any job for the enterprise in the domain of ICT or domains near to it.

1.4 Pitfall 4 If you win in your technology change activities and you rest on your laurel, you will loose. In early nineties, due to the down economy period in the case study country after the political transformation and due to the naive liberalism, the growth of the power system was stopped and even reversed.

Corporation 2 took over the power industry automation centre mentioned above. Obviously, they looked for buildings and equipment, and not for the

85

best experienced staff in ICT systems for the power industry.

The Conference program may be considered as asking the question: Who is a winner and who is a loser in implementation of ICT systems in automation? If the parties of such question are limited to the ICT pioneers in power system automation and the Corporations, the answer is: both. The pioneers lost a chance for continuation of their work and had to find other jobs. And the Corporations:

Corporation 3 established a branch in Wroclaw, employing people of zero experience in power industry automatic control. The same was with Corporation 4 that established a branch near Wroclaw (in Klodzko, Lower Silesia). Though the Corporations tried to operate in the Lower Silesia, the forging shop the computers and ICT applications of the case study country, they employed no one of the local leading designers, the actual pioneers of ICT in the country.

Corporation 1 could not find any work for the exbiggest Manufacturer of ICT systems in the case study country. In a year or so they started a program of voluntary quitting the job by the employees and soon decreased the employment from some fifteen hundred to two and half hundred and, after a few years, completely sold the enterprise.

The methods used by the Corporations may be presented on the basis of experience of the most successful leading designer of ICT implementation of the power industry

Corporation 1 lost quite a lot of money and prestige, at least in the Lower Silesia and the case study country.

The designer submitted a proposal of work for Corporation 1. In two days, he received a rather long e-mail from the Corporation director for staff, explaining that they had studied his website very carefully and that he will never get any job from them. So he was on their black list. He looked at his website once again; everything was normal: projects, hardware, software. Only one thing was rather odd: he had led the project of Tv Solidarity in the middle eighties. Is it possible that he has been put on the black list for the anti-communist activity in contrast to the people put on such lists by McCarthy because of their communist activities?

Corporation 2 did not get a technical nor economic success and, after a few years, had to wind up the power industry automation centre in Wroclaw and to move, almost completely, from the Lower Silesia. A loss of money and prestige was the overall result. Corporation 3 also did not get a success in the Lower Silesia. So they decided to close their Wroclaw branch in an illegal way, however. They simply declared closure without any negotiations with creditors, required by the valid law. A loss of money, lost prestige and an opinion of a plain thief were the results.

Corporation 2, though operating in the same domain he had been the definite Polish pioneer in before, never proposed him any employment in ICTs in the power industry, for instance as a consultant. On the contrary, as he was also a professional translator, the Corporation that ordered translations for their localisation purposes, requested from the translation offices that the designer might not be a subcontractor. This looks rather ridiculous especially when it is remembered that he has been the pioneer in ICT applications in power system automation and he and the teams he has led have been responsible for devising of local terminology in that domain.

Corporation 4 suffered the lowest losses. However, for a long time period they could even pay wages for their employees in due time and had to move out, partially, from the Lower Silesia. Considering the results of both parties, a question may be asked: is there any sense in loose-and-loose games, when everybody looses? 1.6 Pitfall 6 Insufficient ICT transfer control by the authorities results in rather poor effects.

Corporation 3 commissioned quite a lot of translations to him till they learned his professional carrier. Then they stopped to do so and even did not pay quite a serious amount for his earlier translation work. Corporation 4 behaved similarly, though they were honest enough to pay all their dues.

State

During the period of vast ICT transfer to the case study country, local enterprises had to compete alone, without almost none support from the government, with big Corporations of well developed countries. And they lost. The biggest manufacturer of computers, a big automation enterprise and the power industry automation centre mentioned above, etc. were closed down. The State did nothing to protect them. Such naive liberalism may lead to the situation similar to that of the case study country, where the government permitted for that, at the end of the

The methods used by Corporations 1-4, i.e. negative mobbing, black lists, direct stealing, are commonly recognised unethical, immoral, non cultural and socially harmful. And, from the point of view of the Corporations, the methods proved to be ineffective, since the all four Corporations lost and had to move out from the Lower Silesia.

86

XX-th century, the biggest manufacturer of computers in the case study country was liquidated.

ICT and automation transfer in the case study country. The authors dedicate the papers to the actual pioneers of ICT and automation in the country.

1.7 Pitfall 7

The authors are not magi and can not improve the fate of the pioneers. But there is small chance that somebody reads the papers and protects himself or herself against inappropriate technology transfer ways. People, especially young ones, usually look at big corporations and their management through the perspective of their technical and financial success, and think they are wise. That is wrong: they are educated only, and education is not a synonym of wisdom. To be recognised wise, they must actually understand and accept the basic human notions of ethics, morality and culture, and not to try to make any harm to other people, especially weaker ones.

The interrupted experience continuation is a severe inhibitor of the ICT change. The very basic reason of the failure of the case study country enterprises involved in the ICT change and the Corporations 1-4 was that the actual local experience in the domain was not continued nor enlarged. This would require, to make use of the knowledge and experience of the successful leading designers of ICT projects. They knew the local conditions in the highest degree and proved that they could cope with them.

In some aspects, the papers touch dirty things and people usually are ashamed of writing on such things, especially for technical and research events. However, the authors do think that that it is beneficial for people doing dirty tricks. If so, the experiences of the pioneers will not be lost uselessly.

Changing them by other people, even those of knowledge and experience for other countries, proved to lead to rather unnecessary losses. Some corporations claim that they do continue the experience of the acquired enterprises: the managerial experience. Indeed, the Corporation 1 maintained the ex-director of Elwro on his position and did nod do any bigger changes in the lower level management.

ACKNOWLEDGEMENTS The authors gratefully acknowledge the support of Marion Hersh, Larry Stappleton, Frederick Kile and Peter Kopacek in publication of the three papers.

This seams rather ridiculous. The design staff, especially the leading designers, were the people who successfully implemented the pioneering ICT systems in the case study country, often against rather severe resistance of the managerial staff. And no of the corporations mentioned above even tried to find them. On the other hand, the managerial staff were directly responsible for the difficulties of ICT and automation enterprises; in general, they were poorly educated and experienced to fight effectively for their enterprises in the times of international markets, free competition, PC, etc. But may be it was a reward for them: for destroying their enterprises, they were offered to keep their positions.

REFERENCES Hersh M.A., Technology Change, Technology Transfer and Ethics: Part I, In: Kopacek P. (Editor), The 8f IF AC Conference on Social Stability: The Challenge of Technology Developments, Vienna, 2001a. Hersh M.A., Whistleblowers - Heroes or Traitors?: Individual and Collective Responsibility for Ethical Behaviour, In: Kopacek P. (Editor), The 8f IF AC Conference on Social Stability: The Challenge of Technology Developments, Vienna, 2001b. Jancev M. and Cernetic J., A Socially Appropriate Approach for Managing Technological Change, In: Kopacek P. (Editor), The 8th IFAC Conference on Social Stability: The Challenge of Technology Developments,, Vienna, 2001. Kile F., Insights into International Social Stability, In: Kopacek P. (Editor), The 8th IFAC Conference on Social Stability: The Challenge of Technology Developments, Vienna, 2001. Lewoc J.B., Rozent M., Saczuk I.., The Computing Power and the Computer Systems for the Power Industry in Poland, In: MICC (IFAC), Praha, 1992.

2. FINAL REMARKS The purpose of this paper is not to fight for improving the situation in the case study country. This milk has been spoiled already, the pioneers of ICT implementation in power system automation in the Lower Silesia found new jobs, the leading enterprises were closed. The actual purpose is to present the problems to some other people; perhaps they can protect against the adverse effects described. Since the national poet of the case study country wrote some time ago: The one whose dead corps may create a step to the town of glory for other people is also happy. The paper is the third and final of a series of papers devoted to the ethical, moral and cultural aspects of

87

This page intentionally left blank

Copyright © Improving Stability in Developing Nations through Automation Prishtina, Kosovo, 2006

ELSEVIER

IF AC PUBLICATIONS

AUTOMATION, CONTROL AND SUPERVISION OF COMBINED HEAT AND POWER SYSTEMS Jozef B. Lewoc1, Antonina Kieleczawa2, Antoni Izworski3 and Slawomir Skowronski.3 Leader (Leading Designer), ul. Powst. SI. 193/28, 53-138 Wroclaw, Poland IASE (Institute of Power System Automation), ul. Wystawowa 1, 51-618 Wroclaw, Poland 3 Wroclaw University of Technology, Wyb. Wyspianskiego 27, 50-370 Wroclaw, Poland Corresponding author: Jozef B. Lewoc, ul. Powst. SI. 193/28, 53-138 Wroclaw, Poland; tel: +48-71-361-4436; fax: +48-71-783-789-7; e-mail: [email protected] Abstract: The paper presents some technical aspects of the automation, control and supervision of combined thermal and electric power in Poland, on the example of major products and implementations developed by the Institute of Power System Automation during several previous years. The general diagram of the hardware system Powerster designed and developed for combined heat and power plants is discussed, together with the individual solutions on the three hierarchy levels employed. The system software (applications and middleware) developed earlier is also depicted. On the example of the system Powerster, some positive aspects of the technology transfer process based on transfer of hardware to the developing countries and the use of local experience are specified. Copyright © 2006 IF AC. Keywords: heat and power systems, monitoring, supervision, computer networks, technology transfer.

1.

However, some of them, who stayed with IASE or co-operated with BPBiT Leader (a small company continuing the Polish pioneering design and research work in ICT and automation), could undertake the challenge released by the Polish thermal power industry in the late nineties and design and develop the most severely needed systems.

INTRODUCTION

The Institute of Power System Automation (IASE), Wroclaw, Poland, have been a definite leader in design, development and implementation of Polish pioneering monitoring and control systems for the power industry (Lewoc, 1992). Since the seventies, IASE have educated a wide experienced staff for design of the system functionality, selection of the hardware and software structures, design and development of specialised hardware and software modules (including single-board microcomputers), design, development and implementation of complete systems, training, etc.

This challenge is particularly sever in the times of the energy crisis endangering the political and environmental stability. It needs to look for new renewable energy resources. But a lot can and should be done in the area of energy conservation and the paper is devoted to this domain.

Due to the reasons described by Izworski, 2001, IASE suffered severe financial problems and most of the experienced staff left the Institute during the very low economy period in Poland in early nineties.

It should be emphasised that after the political transformation of 1989 and after the earlier Polish down economy period, the demand for electric power decreased considerably in Poland instead of pre-

89

planned continuous increase. Therefore, IASE, the biggest organisation in Poland developing computer systems for the power industry, had to look for a market for themselves in the thermal and electric power industry.

2.3. The Operator's Level The hardware basis of the Operator's Level consists of the Processing/Graphic Station (PGS) PC's (preferably in the industrial version) co-operating with colour visual displays (including so called graphic walls (Synelec, 2003)), printers, standard and and/or specialised keyboards and manipulators. The PC's are synchronised with the satellite Global Positioning System (GPS) and interconnected one to another and with so home-made gateways.

The paper will describe the basic product for the application range, the Powerster system, based on the earlier IASE product, the Master System. A typical structure of the Powerster system for a combined heat and power generating plant, EC Wroclaw, will be discussed. Some issues connected with the technology transfer process will also be addressed.

The complete structure operates on the QNX platform, under control with the home-made middleware and software. The middleware controls the basic system functionality covering communications, processing, archiving, presentation and man-machine dialogue.

2. THE POWERSTER SYSTEM 2.1. General A basic problem needing solution in combined heat and power generating stations (HPGS) is connected with monitoring and control of the Heat and Power Generating Units (HPGU) operation. Due to the huge number of analogue and digital variables (both primary and secondary (calculated variables)), this requires application of advanced computer systems/networks.

Supervision and Management Level

The first Polish first system intended specifically for monitoring of HPGU's was developed on the turn of the eight decade. The schematic diagram of the system which was improved in compliance with the technology transfer process in Poland is shown in Fig, 1.

Gateways (GATE)

Operator's level

2.2. The Process Level

Operator's stations

The Process Level (PL) is constituted of the Intelbased microprocessors (designed and developed by IASE and called the cells), the analogue and twostate Input/Output modules, the communication module (MasterNet) being a redundant version of CSMA-CD (Ethernet) LAN) and the power supply circuits required (grouped in so called automation hubs). The LAN MasterNet interconnects the process stations with PL and the Operator's Level (OL) as well as the Engineering Subsystem Level (ESL).

x1 Si

"S \S

Communication / processing stations

Engineering subsystems Configuring / diagnostic st.

Process level

The PL functionality, realised under control of the software platform (home-made operating system) and of the applications software written under the specialised, object-oriented programming language REGIN-H, and with use of another home-made program, the graphic editor, covers the following basic tasks:

Process station

Field Ins/Outs Sensors & actuators

the user algorithm executed in a cell, transmissions between the cells, exchange of data with OL, buffering of event messages at the level of the cell and hub.

Fig. 1. Schematic diagram of Powerster

90

The application software covers a set of customisation parameters (process variable catalogues, definitions of screens, reports, calculating algorithms for secondary variables, etc.), prepared in the Configuring/Diagnostic Stations (CDS) and controlling the computational and presentation tasks on the OL.

Visualisation, automatic and direct control, realised with specialised screens (Scheme, Events, Alarms, Catalogue, Plot) and windows: Automatic Control, Direct Control, dialogue and information boxes, Data-logging and reporting (many report templates and home-made report writer, uninterrupted logging on magneto-optic compact disks, colour screen hard copies, etc.), Operator-machine dialogue (via standard and/or specialised keyboards, visual displays (including the Graphic Walls) and manipulators).

The OL functionality covers, among other things: Communication with PL (standards available: Ethernet, RS-232, RS-485), Automatic archiving of all variables defined in the system, Event archival (normally, ca. one million events), Secondary variable processing (performed with use of more than 30 standard computational functional modules),

A typical plant scheme is shown in Fig. 2.

150

TUM Fig. 2. Scheme screen Secondary (functional) processing editor (more than 30 standard algebraic and logical functions, expandable of special functions developed on the Customer's demand), Screen editor (to create screens with plant synoptic schemes, bar-graphs, plots, colours under control, tables, digital values, logic data, etc.), Report writer to create report templates for the field and secondary variable data, Field signal and automatic and digital control structure processing editor (to create the algorithms with a graphic environment.

2.4. Engineering subsystems The Engineering Subsystems (ES) enable to develop Powerster software on the basis of customisation, thus facilitating the task and avoiding many possible programming errors. The customising functionality is realised by the following editors of ES: Catalogue editor (used to create the sets (called catalogues) of the field variables, secondary variables, operator's variables, automatic control systems and direct control systems),

91

2.5. Supervision and Management Level

the total plant. For this purpose, the HTTP processes in the gateways may transfer data to the WWW servers (basically) and to the other SML level subsystems. In addition, it is being investigated if it is feasible to employ at SML some specialised expert systems, e.g. G2 (Gensym, 2003), intended towards improvement of the working media production profitability.

It is planned that the Supervision and Management Level (SML) will enable to develop the Computer Integrated Manufacturing and Management (CIMM) systems considered (by the engineering and managerial staff of heat and power generating plants and other works (Franasik, 2001, Izworski, 2006)) as the necessary component for optimum operation of

Supervisory &]Vknagpment Network

HPGS Control

Boiler 1 operator's computer

Turbine 1 operator's computer

Turbine 2 operator's computer

Boiler 2 operator's computer

Water boiler operators' computers

Operator's Level (CL)

Process Level (PL) Process Station 1

Process Station k

1 Heat and power generating [dant under monitoring and control

Fig. 3 Schematic diagram of exemplary solution

92

The basic aspects of the technology transfer in the domain of the power industry computer automation systems are discussed by Lewoc (2005).

3. EXEMPLARY PROBLEM SOLUTION An exemplary solution of the monitoring and supervision problem will be described hereinafter for the case of a combined thermal and electric power generating plant, supplying a several hundred thousand inhabitant town in Poland. The solution is presented in Fig. 3.

The successful solution described hereinabove and the other ones were implemented in spite of that, in the nineties, during the time of the naive liberalism in Poland, IASE had severe troubles connected with their existence. However, the staff educated by the Institute was so numerous that the rest that stayed with IASE were enough to undertake the new challenge of the power industry and, even more, of the overall economy of the country. It worked even though, because of the financial problems, the financing necessary for novel solutions was severely limited and, even more, many possible customers accepted the idea of numerous High-Tech providers that in the technology transfer process, Polish engineers should be treated as lay ones only.

The plant under control consists of two HPGU's (100 MW each) and two old water boilers which are in operation in order to enable fast changeovers of the working HPGU's. In total, the plant under monitoring and supervision is described by several thousand two-state and analogue variables (primary or secondary ones). The process stations (automation hubs) intercommunicate with the OL computers via the LANs at the levels under discussion (and relevant gateways).

And even the Western investigators promote other ways of the technology transfer (Hersh, 2001; Kile, 2001). This paper presents a success of the IASE approach to the technology change/transfer process.

At the Operator's Level, there are two computers per a HPGU (one for the boiler operator and the other for the turbine operator). In additions, there have been provided two computers for the water boilers and one for the overall plant Control Engineer.

The basic reasons for this success were as follows: • Prospective employers in Poland (and in other developing countries) are stopping to believe in honest intentions of the big HighTech providers, • Experienced Polish designers understand better the needs of the Polish heat and electric power industry, • The problems occurring in the developing countries are often too difficult for designers of the developed, "well-ordered" countries, • The presentation layer is of high importance in the monitoring systems; therefore the thoughtless localisation may result in rather inefficient systems, • A heresy may be repeated here: a successful designer in a poorly developed country is better than his or her High-Tech country counterpart, since he or she has had to work in much more difficult conditions for the success.

The graphic wall shared by the authorised operators/engineers enables large-scale graphical display of the data acquired and control of process actuating members. It is very useful for displaying of important plant schemes, especially in emergency conditions when the operators and control engineers should have at their disposal some scheme and other display forms of high readability, reliability and robustness. The OL computers intercommunicate with the Supervisory and Management Network (SMN) ones. However, it should be noticed here, that the staff operating at this level are not allowed to perform direct control on the plant (for such actions, only the Control Engineers and Operators are authorised); the customers at this level may only select the data they need for their work. The solution was implemented in 2002. Usually, the benefits generated by similar monitoring and supervisory systems in the heat and power industry are assessed as some dozen or so percent of the thermal and electric energy sold by the plant. At this assumption, the system paid back many times during the first half year after being commissioned.

4.2. Negative examples The above theses may be also supported by the following negative examples. Examples of Elwro. The problems experienced by a big Western corporation and Poland after acquisition of the ex-Polish biggest manufacturer of computers, Elwro, are described by Izworski, (2001). Let us describe here in some more detail another project.

4. SOME TECHNOLOGY TRANSFER ISSUES

The same corporation was granted a contract for design and development of the operator's systems for a 360 MW Power Generating Unit (PGU). They

4.1. General issues

93

subcontracted the order to a local power industry automation company, established on the basis of the team educated in late eighties on the project intended for the same power plant (and abandoned due to the down-economy period in Poland). However, this company did not employ the most experienced designers of the team. Moreover, they enlisted the most experienced leading designer on their (political) black list. The result was that the task was not executed in a proper way. The basic mistake of the corporation was that they did not look for the best designers available but for the managerial staff, buildings, equipment and so on. Perhaps the things are of some importance, but not in the case of the technology transfer where they should be substituted by the High-Tech ones.

recognise the technology receivers as lay people sets only. They should seek for the most experienced designers available in relevant countries and employ them to develop the best solutions for the countries under consideration. They should not base the technology transfer process on the thoughtless localisation only. In the other case, they will loose sooner or later. In addition, such approach is simply unethical and cultureless, especially in the third Millennium. And, paradoxically, in a longer perm, such companies loose (their image and money). REFERENCES Franasik L. et al. (2001), System Media - A Step towards Computer Integrated Manufacturing and Management Systems, AMSE Press, pp. 37-46. Gensym (2006), G2 Expert System, Available at: www.gensymcom/index.html. Hersh M.A., Technology Change, Technology Transfer and Ethics, In: Kopacek P. (Editor), Social Stability: The Challenge of Technology Development (SWIIS '01, IF AC), Vienna, pp. 6368. Izworski A and Lewoc, J.B. and Piwowar B. (2001), Some Aspects of Technology Transfer - A Case Study, In: Kopacek P. (Editor), The Challenge of Technology Development (SWIIS, IFAQt, Vienna, pp. 55-60. Izworski A. and Lewoc J.B. (2003a), Robustness Comparison of Enterprise Energu Distributions Systems of Various Topologies, In: Robust Control Systems (IFAC), Milan. Izworski A., Lewoc J.B. and Skowronski S. (2003b), Robust Performance Case Study: Topology of System Media, In: Control Applications of Automation (IFAC), Visegrad. Izworski A., Lewoc J.B. and Skowronski S., (2006), Development of computer integrated manufacturing and management systems, In. Information Control Problems in Manufacturing (INCOM2006, IFAC), Kile F. (2001), Insights into Future International Social Stability, In: Kopacek P. (Editor), Social Stability: The Challenge of Technology Development (SWIIS '01, IFAC), Vienna, pp.4548. Lewoc J.B., Rozent M. and Saczuk I., The computing power and the computer systems for the power industry in Poland, MICC (IFAC), Prague 1992. Lewoc J.B., (2005), Practical Ethical Aspects for Technology Transfer and Change in Computer Control System Development, In: The 16-th IFAC Congress, Prague. Synelec MT (2006), Graphic Wall, Available at: www.synelec.com/corporate.html#.

Another example. A big Western corporation acquired the Power Industry Automation Centre that was the entity that had developed, in co-operation with IASE, the system MASTER that had been an predecessor of the system Powerster described in the present paper. However, they did not take care for co-operation of the most experienced staff available in Wroclaw. Even more, they put the name of the most experienced leading designer on the (political) black list, like the first corporation did. In addition, they forced the local staff to localise thoughtlessly their systems. They lost and had to liquidate the Power Industry Automation Centre in Wroclaw. Still another example. A big Western corporation established their branch in Klodzko, Lower Silesia, Poland and treated their employers as lay ones only. They never looked for experienced Polish designers for the power industry automation. They lost and had severe economic troubles. Some people present opinions defending the big Tigh-Tech providers: the recession. However, the authors believe that the more severe is the recession, the smaller mistakes can be tolerable. And the big providers mentioned above (in practice, all operating in the Lower Silesia region, the ex-kingdom of the Polish heat and power industry automation and computers, in general) made the biggest, philosophical, the very basic mistakes The negative examples presented hereinabove do confirm the thesis specified at the end of the previous Subsection 6.1 and the more general ones defined by Hersh (2001) and Kile (2001). 5. CONCLUSIONS In the technology change and transfer process, the developing countries have to take care on development of their own experienced staff. The High-Tech companies working for the heat and electric power automation industry must not

94

Copyright © Improving Stability in Developing Nations through Automation Prishtina, Kosovo, 2006

ELSEVIER

IF AC PUBLICATIONS

COORDINATION OF THE MULTI AGENT SYSTEM AND ITS APPLICATION TO ROBOT SOCCER Arbnor Pajaziti1^ Ismajl Gojani1^ Gazmend Shala1^ Peter Kopacek2) ^Mechanical Engineering Faculty, University of Prishtina, KOSOVA [email protected], [email protected] , [email protected] 'Institute for Handling Devices and Robotics, Vienna University of Technology, Austria kopacek@ihrt. tuwien. ac. at

Abstract: Multi-Agent-System (MAS) consists of a group of agents that can take a specific role within an organizational structure. The coordination of the multi agents is associated with a lot of supplementary work in advance. The robot soccer system is a good technique for multi-robot cooperative autonomous systems, with highly coordinated operations and movements as to fulfil specific objectives, even under adverse situations. Robot soccer system helps to understand and provide with theory and principles for constructing complex systems with multiple agents and their coordination or competition in dynamical environments. On the other side robot soccer is a good tool for the edutainment of technical fields such as mechanical, electrical, software engineering etc. In this article, we introduced the development of the robot soccer system at Mechanical Engineering Faculty. According to the experience with the students the robot soccer is very good tool for the education of different engineering as well as technical subjects. They can study and perfect their knowledge on simulator application and its real time application to robot soccer that can enable to implement their own ideas, in particularly. Key words: Multi Agent System, Robot Soccer, MIROSOT, Image processing, Control

robot it is allowed only wireless communication. Mobile robots should act under the unknown environment with static and dynamic obstacles.

1. INTRODUCTION Robots mainly in industry are used. Today there are new fields of robot applications, for instance: robots for service works, in house, restaurant etc. In future, personal robots will have several advantages, as is decrease of the service costs, which reduces the needs for the workers capable for simple and dangerous work routines (Kim et al, 1998a, 1998b; De Almeida, et al., 1998).

2. ROBOT SOCCER SYSTEM Beside the severity of hardware as is CPU (Central Processing Unit), actuator, other sensors, the robot soccer system is distinct in control algorithm, strategy and complete integrated system. The hardware and software selection among different methods is a challenge for robot designers. Robots supplied with visual system, main computer as well as communication system, are a need to perform the robot soccer game (Figure 1).

We have chosen two methods of robot movement in coordinated manner, first method is the simulator, while the second one is that through computer to realize the movement up to real robot (Han, et al., 2000; Kopacek, 2000). Illustration of mobile robots that play football game was supplied with motors and drives, at two wheels and commanded by computer has been given in this paper [6]. Each robot should be fully autonomous, supplied by motors and power transmission. At each robot is installed the processor. For each interaction between main computer and

There are two operation methods of robot soccer in use: • Robot soccer system based on vision, Robot soccer system based on mobile platform.

95

The first is robot soccer system without memory in robot and the second is robot soccer system with memory in robot.

relatively complicated. Robots can use double RF digital communication with FM module. The architecture of robot soccer is a fully autonomous system that consists of determined cycle and processed with high requests that can be involved as following: • CCD camera installed above the football playground receives two dimensional (2D) pictures and sends those as unprocessed data to the picture's receiver, Picture's receiver installed on PCI or on the AGP angle of computer digitalizes twodimensional (2D) pictures received from camera, The software that processes the picture works directly with picture's receiver and its task is to analyze the digitalized image, to explore and detect the position and orientation of objects. For this task, there is a group of algorithms of detecting and filtering, because there is a need for processing speed and detection control for the robot soccer games. The detected objects are ball, robots of home and opponent team, their positions and orientations sent to judged module, Judged module is place where the strategies took place and the commands are spreader to each robot. When the commands are settled, they are sent to the communication module, Communication module transforms to desired position for each robot, to transmitted commands to robots. Transmitted information is velocity and angle that move the robot, but not the commands for each wheel, Robot hardware receives the motor commands, decodes information and commands wheels to execute the movement.

Figure 1. Robot soccer system with visual system, main computer and RF communication system The useful method but not right one, is the usage of visual camera, (Figure 2). Each robot could have personal encoders enabling the calculation of its instant position. The lateral infrared sensors (IR) are useful for detection of obstacles. Differently, from the traditional method of light quantity measurements, in order to make it desensitize under the environment light conditions, the module transmission signal are used.

Figure 2. Analogue colour CCD camera

3. MICROROBOT ARCHITECTURE OF ROBOT SOCCER The design philosophy is to realize the autonomous micro robots with different functions; so they could be used only to soccer games, but also to other applications. Robot soccer used in MiroSot mainly consists of PCB card with equipment installed on surface, one body coverage for actuator security purposes for movement and battery for power. Central Processing Unit (CPU), lateral infrared sensors (IR), contact sensors and moving mechanism enable the robot movement control during the obstacle avoidance and other intelligent operations. Robots should carry the energy source and rechargeable batteries. Energy characteristics, competition time duration, size and weight are main factors in battery selection.

3.1. Main computer The main computer task is to: ft Execute algorithm for visual system (by finding position of ball and robots, sending coordinates and others) ft Generate command of movement based on strategy and ft Communicate with players.

3.2. Communication module

According to MiroSot regulation robots should not be connected with wire, therefore the separate IR or RF digital communicated control system is needed. At control method through IR (wireless) the circuit is relatively simple when the number of channels is less, but when this number is increased, it becomes

RF semi-duplex modules (frequency: 418, 433, 869 or 914 MHz) are used for wireless communication between robots and main computer. Currently there is in use the 9600 baud velocity, which can be 3 times greater, but with risk of error transmission.

96

With baud velocity of 9600 each command consists of 9 bytes each takes 9 (ms). Communication velocity is very important for robot soccer, therefore the Bluetooth model that is commonly used at mobile phones could not be used, because communication velocity is relatively slow - it takes more than 20 (ms).

Coordination of mobile robots in several manners are performed.

A new robot for football at the Institute for Handling Devices and Robotics (IHRT), Vienna University of Technology has been developed later (Kopacek, et al, 2001).

In this paper are mentioned two of them, the first is the simulation software "3D Robot Soccer Simulator", while the second one with software called "Visio". With the programs, "3D Robot Soccer Simulator" movement coordinated of mobile robots is performed through tables prepared in software Microsoft Excel. Regarding the second method of robot coordination, all through programming language C++, when after this coordination connected to "Visio" software where the executed movement, are performed.

ROBY-GO (Figure 3) is a robot that moves with two wheels and is characterised with simple structure, compatibility and modular construction.

4.1. Software for simulation of mobile robots "3D Robot Soccer Simulator "

3.3. Main mobile robot - ROBY-GO

Is a program for simulation of robot soccer that was developed by a group of students from Australia, (Griffith University, Australia). According to MiroSot league categories, game that in this program of the middle league has been played (5 against 5), (Figure 5).

Figure 3. Mobile robot ROBY-GO Robot shape is defined with the following dimensions 7.5 [cm] x 7.5 [cm] x 7.5 [cm] and hose the velocity equal to 3 m/s. The RF Antenna height for communication in the robot size is not calculated. It has two DC motors with PWM (Figure 4).

Figure 5. Working window of the"3D Robot Soccer Simulator" There were two teams, blue and yellow team, that used strategies created before through simple text (Lingo) and C++ programming language. Game rules are the same as in real game of this category. This program was very important, since the strategy before it has employed to real game was tested. Minimal requests for properly work of "3D Robot Soccer Simulator" are: Pentium III 600 MHz 256 Megabyte RAM TNT2 3d Graphics accelerator with 32 Megabyte RAM 24x CD-ROM Screen resolution 800 x 600 16 bit Sound Card

Figure 4. Mechanical structure of mobile robot ROBY-GO ROBY-GO as a robot that plays in MiroSot category, but also for other tests of multi agent systems is not only used (Han, et ah, 2003).

4. MOBILE ROBOT MOVEMENT COORDINATION

97

*

Microsoft Windows 98 DirectX 8.0 10 Megabyte free space in the disk.

^"*

""" (VfririiTil

/^~ 1

ft

-|

" fH 1

FTMTEE15

hm

-

PfKtUck FMHty FnaDd Eld IIEL

1i '

4.2. Robot dancing (Parade) In order to realize the robot dancing (parade), another method (strategy) is used. This method of dancing robots at the IHRT in Austria is developed (Han, 2003). The robot movements are not common, but coordinated. This means when one robot is moving another robot is waiting for the order. Programming also from the students that do not possess the knowledge of C++ can developed. This by the usage of sheet in Microsoft Excel is realized (Figure 6).

mmiaii EJUEIH _.

i

j

-—*

' mM

ft

IBE

Ml

SC4RE

l

IIPLISCOI[

&DF1TUM &H1I

Figure 7. Initial trajectories of robots

There are five different functions: 1. ParadeWait - Robot waits for a proper time (time in milliseconds is measured). 2. Parade Position - Robot moves to certain position. 3. ParadeRotate - Robot rotates to certain angle (in absolute coordinates). 4. ParadeCircle - Robot moves along a certain circle. 5. ParadeTimeRotate - Robot rotates within a certain time.

4.3. Program that enables real movement of "Visio " robots By usage of the program "Visio", one can configure the colours of robots and ball, modification of picture (view), determination of robot size, determination of playground size, robot testing (velocity of each wheel is given particularly), etc. The view of this program is shown in the following picture:

•

H8SH

•: tnii • cat

i,r | MHMIki u M1

Ui

•i.

. . . r r

1

F

r r r r

kali«

IBEJ

* |

• N""*l| ••flj- |

i u ru

HpJ

rjs: rare1^•n

- • ( — :•:-

f1

_ l _ .-II

*•?

r •

r •>

•

>B •

••Ml U *

-

n ••

Figure 6. Excel sheet with strategy for parade Figure 8. Window of the program " Visio' In addition some examples of robot programming coordinated movements are given. Movements of three robots fora short parade are programmed. This kind of programming, in the beginning the first function (ParadeWait) should be written. Then, the function 2 (ParadePosition) has been used, where the initial position coordinates of robots have been given. E-Nr.

R±oti1 F

1 1 2

P1 0

Robcti2

P2 P3 P4 P5 S1 S2 S3 S4 S5 F 0

0

P1

Robcti3

P2 P3 P4 P5 S1 S2 S3 S4 S5 F

0

0

0

0

0

0

0 1

0

0

125 105 0 150

0

0

0

0

0

0 2

125

80

0

4.4. Robot programming in C++

0

0

0

0

0

0

0 1

0 150

0

0

0

0

0

0 2

P1 0

P2 P3 P4 P5 S1 S2 S3 S4 S5 0

125 160

0

0

0

0

0

0

0

0 150 0

0

0

0

0

0

0

According the given coordinates in the previous table, of the software "3D Robot Soccer Simulator", the position of robots is as following:

98

In order to realize the movements of robot soccer game, program that enables movements with program "Visio" has to be connected. Implementation of C++ to connection and compiled modules is used. These modules deals with camera recognition from program, framegraber recognition (module installed in computer that processes the picture), antenna recognition that enables communication between computer and robots, etc. Those modules in program "Vision.dsw" and shown below are placed:

////////////////

ROBOT

SOCCER

5. CONCLUSIONS

PROGRAM

// VisionView.cpp : implementation CVisionView class #include "stdafx.h" #include "Vision.h" #include "VisionDoc.h" #include "VisionView.h" #include <math.h> #include <mmsystem.h> #include "SettingColor.h" #include "SetBoundary.h" #include "SetPixelSize.h" #include <winsvc.h> #include "ddraw.h" #include "Comm.h" #include "Robot.h" #include "Miro03.h" #include "TunningDlg.h" #include "ColorTunning.h" #include "Robot test.h"

Robot soccer is a Multi Agent System (MAS), with highly coordinated operation and movements so to fulfil specific objectives, even under adverse situation. From the scientific point of view, soccer robot is an intelligent, autonomous agent, carrying out tasks with other agents in a cooperative, coordinated and communicative way. On the other side one of the newest trends in robotics is the application of robots in entertainment, leisure, and hobby.

of the

Vision system is the only sensor available to determine the position and orientation of the robots. Without it, the robots would be blind and lost, making their ability to move or play virtually impossible. The entire system is dependent upon the vision system and the information it provides. Besides direct programming with robot soccer, it is possible that through the simulation programming to developed the strategy of robot soccer. Simulation programs that save the time of design of different strategies are developed.

Program "Vision.dsw" with all modules is shown in the following figure.

With simulation is it also possible to design other movements and trajectories coordinated among the mobile robots as is dancing. The determination of proper position of mobile robots as well as the ball is the necessity of the game and strategy. Without this elementary information system will not function and the game will not be realized.

Z* m •

:

•

m

•

!

•

-Ji

•, . 1 . .

_l

•jr> Z>utoiv,KV«| • 0 •jr>

I

•

•

-

.

, iiti:-K4ii

I . •Ji

Comparing with direct programming of mobile robots, the simulation program behaved with small difference, so one can conclude what has to take measure in order to minimize the position error.

• '-' 'jpl - I — r . . . lii - I

-irwa-

y.iiv.-- i;,i!-tJi-i

6. REFERENCES De Almeida A.T. and O. Khatib (1998). Lecture Notes in Control and Information Sciences Autonomic Robotic Systems, Springer London. Fischer G. (2000). Robots in Entertainment leisure and hobby, Diploma work, Vienna University of Technology, Austria. Han, M.W, P. Kopacek, G. Novak and A. Rojko (2000). Adaptive Velocity Control of Soccer Mobile Robots, Proceedings of the 10th Workshop on "Robotics Alpe-Adria-Danube Region - RAAD '00", Maribor, Slovenia, p. 4146. Han M.-W. (2003). Robot Soccer and Edutainment In: Proceedings of the 2nd FIRA Robot Soccer World Congress, Vienna, Austria, 1-3, p. 45-49. Han, M.-W. et al. (2003). Robot Soccer. In: Proceedings of the 12th Int. Workshop on Robotics in Alpe-Adria-Danube-Region RAAD'03, Cassino, Italy.

Figure 9. Program "Vision.dsw" with all modules

Figure 10. Robot soccer team Figure 10 shows the robot soccer team developed at the Laboratory of Robotics of the Faculty of Mechanical Engineering Faculty, University of Prishtina.

99

Kim J.-H.; Vadakkepat P.; Verner, I (1998a). The FIRA Robot World Cup Initiative and Research Directions, Communicated to the Int. Journal of Robotics and Autonomous Systems,. Kim J.-H. et.al. (1998b). A Cooperative Multi-Agent System and Its Real Time Application to Robot Soccer, Accessible from http://www.mirosot.org/. Kopacek P. (2000). Robots in entertainment, leisure and hobby, Preprints of the IFAC Symposium on "Robot Control 2000, SYROCO '00", Vienna, Austria. Vol. 1, p. 319-323. Kopacek, P.; Han, M.-W.; Novak, G. (2001). Control Tasks in Robot Soccer. In: Proceedings of the 32nd International Symposium on Robots - ISR, Seoul, Korea, 19.-21. p. 756-759.

100

Copyright © Improving Stability in Developing Nations through Automation Prishtina, Kosovo, 2006

ELSEVIER

IF AC PUBLICATIONS

HUMANOID ROBOTS FOR HUMAN LIFE SUPPORT Edmund Schierer Vienna University of Technology Favoritenstrasse 9-11 (E325/A6) A-1040 Vienna, Austria

Abstract: Spoken in technical terms a robot may be defined as a machine which includes almost all classical engineering sciences like mechanical engineering and electronics and electrical engineering as well as "modern" fields of engineering like computer science and material sciences. Non-engineering disciplines like psychology and biology become more and more important in designing a robot that will not only look like a human being but also behave natural in its movements and emotional way of acting. Copyright © 2006 IF AC Keywords: robotics, humanoid robots, household robots, human life support

1. WHY SHOULD WE USE ROBOTS? It is obvious that we as human beings has some awkward physical limitations regarding environmental factors like weather conditions, including temperature, pressure, humidity and so forth. In situations where work has to be done out of these limitations robots can do and should do the job for us. There is no need to think about robots only in outer space, there are many places on earth where it is too dangerous for man to work at like in the deep sea, nuclear plants or waste dumps. But there is also a growing need for robots in supporting human beings in their daily life at home or at work. As we all know domestic duties like cleaning, wiping and ironing are extremely boring and cumbersome. At the beginning of the 21st century a new generation of domestic robots relieves us from such unwelcome responsibilities. They render good services precisely, discreet and, above all, 24 hours a day without getting tiered, hungry or spiritless. Besides these manifold household robots are others which act like domestic animals and try to become a member of the family. Since this is a fast growing and evolving part of robotics this paper has its focus on household robots. Assistant robots are the last step in an evolution from the very first electric household appliances of the

1950s. Though more and more powerful, even the most modern types are basically the same as ancestors. Intelligent robots can give people the freedom to use their spare time in a more satisfying way. Such robots are typically mobile and autonomous devices. Unlike the standard appliances in our households robots differ in a way that they may operate independently and without influence from human beings. In this connexion a problem arises: the robot has to move and operate in a changing environment which is made from and for man. Objects normally are not always on the same place (chairs) what makes it very hard and complex for a robot to navigate. To solve all these problems a modern household robot incorporates numerous sensors which introduce a new problem: the balance between cost of production and selling-price.

2. TYPICAL HOUSEHOLD ACTIVITIES 1.1 Vacuum Cleaning. Vacuum cleaning robots were the first ones especially designed and constructed for household applications. Modern types are small and faint enough to work even at night.

101

The very first vacuum cleaning robot was introduced by iRobot, a US company. One of the founders is the well-known roboticist Rodney Brooks and two other experts from the Massachusetts Institute of Technology in Boston.

Fig. 3. Automower in action (Electrolux) Slugbot, a slug seeking and digesting robot is a development of the Intelligent Autonomous Systems Laboratory in Bristol. In its artificial stomach the slugs are transformed into methane and hydrogen. By means of an electrolytic reaction the robot generates electricity to drive its motors and electronic devices. Automatically adjusts for floor type as Ft moves between surfaces

Fig. 1. Roomba has no problems with different surfaces (iRobot). Roomba, looking like a rolling frisbee, moves on a helical line through a room. It takes less than 20 minutes to clean an average sized room. Obstacle avoidance is done by sensors which was actually developed for military demining robots. eVac as a competitor of Roomba costs only 300 US Dollar and has a much better absorptive strength and has the ability to free itself in case it is blocked. A european vacuum cleaner is the RC 3000 from Karcher. It is faint enough to hear music without being disturbed and flat-bottomed to go under almost any furnitures. Sensors detects steps and landings to prevent it from crashes. A special feature is the ability to find its charging station autonomously.

Fig. 4. Very first prototype of SlugBot

1.3 Surveillance and Control Another field of application is surveillance and control. Modern robots are able to recognize unauthorized persons and release appropriate alarm conditions. Other modes of control include detection of gas, water, fire, temperature or other dangerous materials or compounds.

KARCHER

Fig. 2. RC3000 technical specifications (Karcher)

1.2 Gardening Gardening, especially lawn-mowing is a very time consuming job. A problem of the current generation is that the robots move not systematically but randomly what leads to a relatively high energy consumption until the job is finished. A typical lawn-mower robot is Automower from Husqvarna. It moves completely independently and separates the lawn into sub-divisions. The area has to be initially defined by the gardener by mounting a cable.

Fig. 5. Spy-Cye, home-surveillance tool (Probotics)

1.4 Entertainment, Leisure, and Hobby The most interesting area is where robots can care for the aged and young children. Toshiba Corporation introduced the concept of a ,,life support partner" - a sophisticated robot able to deliver human-centric technologies that provide assistance and support the

102

elderly and young children at home and in such public places as shopping malls or offices There are currently two robots available, ApriAlpha™ V3 and ApriAttenda. While ApriAlpha™ V3can distinguish and recognize voices from any direction, ApriAttenda first of all identifies an individual with its visual sensor and highspeed image processing system by instantaneously recognizing the registered color and texture of the person's clothes. Fig. 8. PaPeRo (NEC). Front

Side

"TV remote control 4 signal transmitter Head patting sensor + Sound recognition microphone 2 CCD cameras + LED* In mouth, head, ears and eyes Head stroking sensor • ifcropnoneE for sound direction detection Ultrasonic sensors +

385mnn

rower switch Speakers • Floor sensor UfI sensor

Internet Wire less modem

Fig. 9. PaPeRo's technical specifications (NEC).

TOSHIBA

(Front) ApriAlpha_v3, Apri sharp ear (Back) ApriAttenda

Ifbot is another example of a communicative robot. He is 45 cm tall and weighs 7 kg. He is able to walk and avoids obstacles by means of powerful sensors.

Fig. 6. Toshiba's robots designed as life-support partners for humans. Especially in Japan robotic domestic animals are very popular. Due to the fact that housing space in cities like Tokyo is expensive and flats are very small animals normally are not allowed.

Fig. 10. ifbot is a very smart toy, with advanced communication features (University of Nagoya, Japan). Fig. 7. BN-1 robot cat (Bandai). Yoshihiro Fujita from NEC's Personal Robot Center designed a communicative housemate named PaPeRo (Partner-type Personal Robot). He follows some basic and simple social rules. If he is fondled he gets nice, attentive and co-operative. Otherwise he gets ill-humored and starts grumbling.

3. HUMANOID ROBOTS FOR DAILY LIFE SUPPORT All the robots mentioned above have some advantages in their special field of application. But the world we live in is made from humans for humans. All the processes necessary to keep our daily life at work are designed for human beings with human proportions and human "gripper devices". It is

103

obvious that a robot which was designed for laundryman is not good at washing the dishes. But we are. It seems that our "design" is suitable for almost all tasks. So why not design a robot that looks like a human and behave like a human? There are actually two sets of reasons one might build a robot with humanoid form. Following the arguments of Johnson and Lakoff, then the physical representation of our bodies is critical and fundamental to the representations that we develop and use for both our internal thought and our language. The second reason is that an important aspect of being human is interaction with other humans. For a human-like intelligent robot to gain experience in interacting with humans it needs a large number of interactions. It will be much more easy and natural for humans to interact with the robot in a human like way if it has a humanoid form.

Fig. 12. Hip joint design of LISA (University of Hannover) For our humanoid Archie we developed a joint module which can be used as a standard for all joints on the robot. One design goal was to mimic as close as possible the physical appearance of a human. This requires a high volume to power ratio, that is, high torque in a small volume. This can only be achieved with harmonic drives, as can be seen in Fig. 13.

There are some basic requirements a robot must fulfill to be designated as human-like. Besides a minimal size he has to have a human shape, that is, one head with eyes, ears and mouth as well as two arms and two legs (with an appropriate number of degree of freedom).

Fig. 13. Standard joint module for Archie (IHRT, Vienna University of Technology). On the other hand, a robot needs the possibility to communicate with his human partners in their most natural interface, a spoken language. Commanding a robot via speech is the most intuitive and most easy way. Controlling it by means of a remote control or some sort of menu system is under no circumstances appropriate. Another essential feature in this connexion is speech, gesture and face recognition. The latter is necessary due to the fact that part of our non-verbal communication is done by gestures and mimics.

Fig. 11. left) A typical European human right) "Complexity" of human skeleton A natural walking and arm movements are unalterable requirements for a humanoid robot if it shall be accepted by its users. Since bipeds are inherently unstable systems (Albert, A., 2002), the control of the drive system is a much more complex task than that in a wheeled robot (Putz, B, 2004; Wiirzl, M., 2005).

For Archie we introduced a distributed controller architecture (Fig. 14) with one node for every degree of freedom interconnected by a fast network. To increase the processing power of the entire network additional nodes can be easily added to the network. Special nodes are reserved for vision processing, sound synthesis and speech recognition as well as for sensor processing.

104

4. SUMMARY To be accepted by the public robots have to be human-like in their physical appearance and social behaviour. Future robots will also differ in a few remarkable points from technologies available today. This paper discusses some of the keypoints important for household robotics and describes features a humanoid should have to be able to do the same jobs as his specialized unintelligent ancestors. MCP

VCP

ACP

RCP

SCP

LCP1

LCP2

LAP2

LAP1

o

ooo ooo

o

ICP

OOO

OOO

Fig. 14. Brief overview of Archie's controller network

REFERENCES Putz,

B., (2004). Navigation of mobile and cooperative robots (in German), Intelligent Handling and Robotics, Vienna University of Technology. Wtirzl, M., (2005). A contribution to robotics especially in the field of expert- and multi agent systems (in German), PhD thesis, Intelligent Handling and Robotics, Vienna University of Technology, June 2005. Albert, A., (2002). Dynamisch stabiles Gehen fur eine autonome, zweibeinige Laufmaschine. 36. Regelungstechnisches Kolloquium Kurzfassung der Beitrage, 20.02-22.02.2002, Boppard, S. 40-41.

105

This page intentionally left blank

Copyright © Improving Stability in Developing Nations through Automation Prishtina, Kosovo, 2006

ELSEVIER

IF AC PUBLICATIONS

STRATEGIES AND PREDICTION IN MULTI-AGENT SYSTEMS

Man-Wook Han

Intelligent Handling and Robotics (IHRT), Vienna University of Technology Favoritenstr 9-11, A-1040 Vienna, Austria Email: han(a)ihrt. tuwien. ac. at

Abstract: The main features of Multi-Agent System (MAS) are the cooperation, coordination and communication between agents. One good test-bed for MAS is the robot soccer. The game of robot soccer provides a good opportunity to implement and test the coordination of group behaviour of mobile robots. For the organization of the group behaviour in MAS the strategies of the group behaviour of each agent as well as the prediction of the future behaviour of agents play important roles. In this contribution strategies for a robot soccer system and the prediction algorithm which is applied in the robot soccer will be introduced. In robot soccer the behaviour of each player in own team can be calculated. But the behaviour of opponent robots and the ball can not be exactly estimated. It is necessary to predict the future position of the opponent robots and the ball. This paper will present the strategies for MultiAgent-System (MAS) which is developed by IHRT and an algorithm for the prediction of position of robots and ball. Copyright © 2006 IF AC Keywords: multi-agent system, organization, robot soccer.

1. INTRODUCTION Approximately 15 years ago the keyword "MultiAgent-Systems (MAS)" was introduced. Multi-agentsystems have emerged as a sub-field of AI that attempts to implement both an unified theory and design principles for constructing complex systems with multiple agents and their coordination in dynamical environments. At that time, computer scientists used MAS as an application example of distributed artificial intelligence. Nowadays MAS are closely related to the design of distributed systems for autonomous agents to be used in manufacturing and robotics applications. MAS deal with interaction and cooperation, organization of agents, action and behaviour of each agent. Several years ago robot soccer was introduced with the purpose to develop the intelligent, cooperative multi-robot (agent) systems and as one of

the first examples for robots in entertainment, leisure and hobby. Robot soccer offers a good opportunity to implement and test control and cooperation algorithms of Multi-Agent-Systems (MAS). From this viewpoint each soccer robot is an intelligent autonomous agent. The robot should be in the position of being able to autonomously carry out the given task based on the available information in such environment. To reach this goal the work on the motion planning and control in a fast changing environment including static and dynamic obstacles is necessary. In more complex environment the position detection and calculation using a global vision system take time. It will be very helpful to manage the behavior of agents by predicting future position based on the previous positions.

107

Centralized control does not necessarily imply that each robot is lacking in intelligence. The on-board software of each robot deals with navigation and control. The centralized controller provides only strategy for cooperation and does not attempt to solve the problem in its entirety.

2. STRATEGIES A multi-agent (robot) cooperation system requires complex coordination strategies to gain the full benefit of applying more than one agent to a problem. With mobile robotic environments, this has added difficulties due to the approximated results of both the hardware and extra software required to run the hardware. In a multi agent system environment, it includes communication delays, vision system integrity, inconsistent field conditions, and other robots. Since it has to be accepted that many of these problems are beyond control, a robust multi-robot control system needs to be developed.

3.

MICRO MOBILE ROBOT

Intelligent Handling and Robotics, Vienna University of Technology developed micro mobile robots, like Roby-Go, Roby-Run, Roby-Speed and Roby-Naro.

There are many problems to be addressed when designing a multi-robot strategy for teamwork. By having more than one moving object in the environment adds dimensions of complexity to path planning. The robots can't learn where the obstacles will be and must contend with them reactively while maintaining a longer term strategic plan. As more robots are in the environment, as more difficult it is to develop team strategies. This is contradictory to how a multi-agent system should benefit from added agents. Each agent has a responsibility to carry out its task to benefit the team. When the team doesn't perform well, it can be difficult analysing where the problem is. The agents may have incorrect strategies or the overlying team strategy could be at fault. As a result of this credit assignment problem, many multiagent designs are environment or problem specific and the technique adopted in one environment may not be transportable to another. Therefore, the environment requires careful consideration when designing multi-agent cooperation. Each agent on the field can obtain enough information about the state of the game not to require any extra information from other agents. Adding communication can add delays in agent reaction as they may wait on incoming information before deciding their next move. The subsequent section looks at our approach to cooperation within the multi agent system domain

Fig. 1 . Micro mobile robot - Roby-Speed A soccer robot is basically not more than a mobile platform, which is connected via a radio module to its superior control unit. The electronic is built up in an open architecture. In theory a robot consists of different units fulfilling special tasks. One important unit is the motion unit. The motion unit controls the motors by a desired trajectory. This desired trajectory (as well as other demand behaviours like acceleration and etc.) has to be transferred to the motion unit. For the design of a soccer robot the considerations above mean that it consists of a motion unit and a connection via radio to its superior control unit. Remaining to the described system the radio module should be connected to a microcontroller, which selects and processes the incoming information. Afterwards a bus provides the processed information to the motion unit. The electronic part of the mini robot has a modular and open architecture and consists of a single board for power electronic, communication and a microcontroller (Hager 2003).

The planner exists on the computer as a form of centralised control for the robots. It examines the state of the game from the robot's perspective and makes decisions as to that robot's next action and uses one motion planning techniques to determine locations for the robots to carry out those actions. In this manner, it abstracts away from the normal multiagent approach of having independent agent control. The credit assignment problem are reduced and moved to one entity rather than several. By having one decision-maker, each robot acts as a remote mobile unit that takes commands. By this method, there is no requirement for communication or independent strategic robot control. Each robot is unaware of the other's intentions, but still contributes towards the goal of team.

The task of the microcontroller is to control both DC motors and analyze the radio data. This board is universally useable and in circuit programmable by the serial port. Furthermore it contains a high speed synchronous serial interface, which gives the possibility to connect several microcontroller boards for different tasks. Sensors can be easily implemented in future.

108

The hardware control unit is intended for both robot constructions. With an area of just 36 x 36 mm the board fills only 1/8 of the reference volume and consists of the following components: • XC167 microcontroller from Infineon with internal RAM (8 kByte) and Flash (128 kByte) • Voltage supply by switching regulators with high efficiency • High speed dual full bridge motor driver • Infrared transmission module • Bi-directional radio module for the frequencies 433, 869 or 914 MHz • Status indication by six bright LED's in different colors • Serial synchronous interface for communication • with further modules, e.g. XScale board. (Putz, 2004)

4. PREDICTION ALGORITHMS FOR MULTIAGENT SYSTEM (Kopacek, et al. 2001) Usually a soccer robot is a two-wheel driven mobile robot, whose size may not exceed 75mm in each side of a cube. It consists of two wheels, two DC motors, a micro controller, a communication module and a power supply and others. It is a very good example for a mechatronic system. The behaviour and efficiency of such a robot depends on the mechanical construction, control algorithm, and the performance and accuracy of the vision system. The vision system is responsible for the calculation of exact object's positions based on the pictures from CCD-Camera above playground. The accuracy of the measurement is not exact enough. The measurement error is maximum up to 5 cm. Also the speed of vision system plays an important role. There is time delay between taking picture with camera, image processing, generation motion commands, sending command via radio frequency, reaching the motion commands and moving robot. In the time between taking visual information and carrying out the command the robot is moving. Assume that the robot moves maximum speed 2.54 m/s and the time delay is 30 milliseconds long the difference between robot's position when the camera takes a picture and robot's position when the robot start moving is almost 8 cm. Furthermore the time between new data provided by vision system is not constant. Therefore it is necessary to predict as well as estimate the future position of the ball and robots based on the previous position information. Several prediction algorithms are already reported, like Kalman-Filter, Least mean square algorithms etc.

This means that if there is an estimate 9(t — 1) based on data up to time t-1, then 9(t) is computed by some 'simple modification' of 9(t — 1) . The counterparts to on-line methods are the so called offline or bath methods, in which all the record data are used simultaneously to find parameter estimates. The least squares method can be applied to a large variety of problems. It is particularly simple for a mathematical model that can be written in the form;


y (t) =

where y is the observed variable, 61,62,, , Qn are unknown parameters, and cpi? (p2,.--->(Pn are known functions that may depend on other known variables. The model is indexed by the variable t, which often denotes time. The elements of the vector cp(t) are often called regression variables or regressors while y(t) is called the regressed variable. The problem is to find an estimate 9 of the parameter vector 6 from measurements y(l), ...,y(N), cp(l),..., cp(N). Given these measurements, a system of linear equations is obtained, namely:

=
=
This can be written in matrix notation as:

Now introduce the equation errors as:

s(t) = y(t) -
i=\

where

denotes the Euclidean vector norm. The

V(0) has unique minimum point given by:

6=

(1)

4.1 Recursive least square estimation In recursive (also called on-line) methods, the parameter estimates are computed recursively in time.

This form of the least squares estimate can be rewritten in the equivalent form:

109

e=

N

4.2 Estimation of the ball and robot

N

(2)

The position information detected by the vision system are inaccurate and old therefore the transmitted position information have to be filtered and the current position estimated. In order to filter the transmitted position information and estimate the current position there is used the recursive least squares algorithm for each coordinates values by using a polynomial of first order for the ball and of second order for robot. For the x values:

i=\

In many cases cp(t) is known as a function of t. Then (2) might be easier to implement than (1) since the matrix O of large dimension is not needed in (2). Also the form (2) is the starting point in deriving recursive estimates. The solution in (2) will now be rewritten into a recursive form. Introduce the notation

x(t) = a

ct

-i-l r

the system is described by the following equation: s=\

s=\

Since trivially

1 to

tl

1 U 1 t

t\

a

tl * b \ c tl

n

it follows that the least squares estimate is given by:

x0 —

_xn_

whereas: t-\

6{t) = Pit)

to=O S=\

tx =

T

eit -1)+P(t)
U = At, + At1 /v

t3 = At3 + At2 + Atx

where 9(t — 1) denote the least squares estimate based on t-1 measurements. The results are quite appealing. The estimate of 6 at time t is equal to the previous estimate (at t-1) plus correction term. The correction term is proportional to the deviation of the

i=\

/v

'predicted'' value 6(t — 1) from what is actually observed at time t, namely y(t).Then the recursive least square algorithm for estimation is given by:

=Tn~ Tn_ n x by this all coefficients Atn are negative. In each step the lines are shifted down with adding Atj to the time values. The last measured x-value is x(0). The same value one second later is x(l). That means the algorithm is independent from the global time. The same applies to the other coordinate values. To solve the equation (4) the recursive-least squares algorithm (3) is applied. In order to obtain the initial values for

0{t) = Bit -1) + K(t)s(t) K(t) = P(t)cp(t) =

(3)

T

y(t)-(p (t)e(t-\)

Here the term s(t) should be interpreted as a prediction error. It is the difference between the measured output y(t) and the one-step-ahead

prediction y(t 11-\\6{t-1)) = (pT {t)9(t-1)

(4)

9, P and R the least squares algorithm is used and after that recursive least squares algorithm.

of

y(t) made at time t-1 based on the model

5. CONCLUSIONS AND FUTURE PERSPECTIVES

corresponding to the estimate 9(t — 1) . If s(t) is small, the estimate s(t) is 'good' and should not be modified very much. The vector K(t) in (3) should be interpreted as weighting or gain factor showing how much the value of s(t) will modify the different elements of the parameter vector.

In this paper the strategies for the Multi-Agent System and the prediction algorithm are presented. The estimation could be more exact when the accuracy of vision system is increasing.

110

To increase the computing power a small microprocessor (DSP) will be added to the minirobot. By the use of the computing power and the multimedia functions of such a small module, there is no problem to integrate the strategy or more sensors like a small camera, a compass, etc. This effort is useful to make the robot autonomous and increase its "intelligence". It is our opinion that this is the right step into the future, because the problem of vision based Hardware-MAS is actually solved.

REFERENCES Ferber, Jaques (1999). Multi-Agent Systems: An Introduction to Distributed Artificial Intelligence, Addison-Weseley, Longman Hager, A. (2003). "Improvement of the mechanical design of mobile mini robots" (in German), Diploma Thesis, IHRT, Vienna University of Technology Kopacek, P.; Aleksic, K.; Han, M.-W. (2001): Prediction Algorithms for Multi-Agent Systems. In: Proceedings of the 10th IFAC Symposium on Information Control Problems in Manufacturing -INCOM2001, 20.-22. September 2001, Wien, p. 153-157. Putz, B. (2004). Development of the new Soccer Robots "Roby-Speed" and "Roby-Naro, "CLAWAR /EURON/IARP Workshop on Robots in Entertainment, Leisure and Hobby - ELH'04 Proceedings", Intelligent Handling and obotics, Vienna University of Technology, Austria, ISBN 3-902161-04-3. Wiirzl M. and E. Schierer (2004): Strategies for Mirosot; Preprints volume of "2004 KoreaAustria Joint Seminar on Intelligent Robotics", 44 - 49 Busan, Korea; 10-28-2004

111

This page intentionally left blank

Copyright © Improving Stability in Developing Nations through Automation Prishtina, Kosovo, 2006

E L S EVI E R

IF AC PUBLICATIONS

Panel Discussion

Control Strategies for a Sustainable Future H. Erbe l\ R.Genser 2), M.W. Han 2), M. Hersh 3) , P. Kopacek 2), L.Stapleton 4)

^ Technical University of Berlin, Germany Intelligent Handling and Robotics, Vienna University of Technology, Austria 3) University of Glasgow, Scotland 4 ^ Waterford Institute of Technology, Ireland

The session commenced with a review of the broad range of serious problems facing humanity. Many of these problems can be addressed using control theory and are affected by a range of different automated and semi-automated systems. Systems approaches can help us to understand both the processes and underlying problems, as well as the interrelationships between different problems in the complex underlying systems. Modelling can be used to understand regulation and control issues and consequently to describe a possible approach to system stabilisation. The types of control strategies required for an analysis of human systems are in many cases different from those used in industrial engineering. Open systems approaches which have been used to illuminate, for example, technology-driven innovation, were discussed. Typically open system models can cope with both the quantitative and qualitative data needed to describe and understand human activity systems. They can then be used to derive control strategies for human activity systems. The world is currently facing a number of very serious crises which could lead to potentially catastrophic instability. These crises include the following: • Environmental crises, including global warming, destruction of the ozone layer, acid rain and atmospheric and water pollution. • Social crises, including serious poverty with half the world's population living on less than the United Nation minimum to meet basic needs of $2 a day, widespread human rights violations, discrimination, injustice and unequal access to the world's resources. • Resource crises, including energy and water shortages and an imbalance between population, consumption and resources, so that human activity has outstripped the planet's source capacity to provide resources and sink capacity to absorb wastes. • Political crises, with excessive influence by the industrial-military complex and the one remaining superpower, which seems to have embarked on a policy of invading other states for reasons other than the officially stated ones. This list is by no means exhaustive. One of the important questions for the control community is the role of control and systems approaches and trying to resolve these and other potentially catastrophic problems in order to stabilise the system. One way to approach this problem is to use multi-loop action learning, a model of which presented below. This model involves the addition of quadruple loop action learning to existing methods. It therefore considers the levels of the individual, the organisation and the wider society. However, further levels could be added to give greater specificity about the wider society by representing, for instance, the local, national and global contexts.

113

wider society

c>rgani!national

societal values

individual values

val ues

organisation

J

s

individual

J

actions '

decisions

evaluation

\

win - win learning

win - lose learning double loop dialogue learning triple loop dialogue learning quadruple loop dialogue learning

Fig. 1 Multi-Loop Action Learning One of the important features of multi-loop action learning is that it has both forward and feedback paths. The forward path shows that actions at the individual level are sited in and influenced by the organisational and wider (local, national and global) contexts. In addition, actions by individuals have a feedback effect and can therefore influence their organisations and the wider (local, national and global) contexts. Similarly actions by organisations have a feedback effect on the local, national and global contexts. This means that change is possible, but not that it is easy, due to pressures to conform on the forward path. The model also shows the value of individuals wishing to achieve change and social stability joining together to form organisations. In this case the organisational context is supportive of change towards social and environmental and social stability and this has a positive effect on the individuals. Similarly the feedback path from the individual to the organisation has a reinforcing effect. These effects have some impact on countering the ethos of the wider social context and supporting the individuals and organisation in working for change. Thus one of the important roles of systems and control techniques is in analysing and modelling complex problems and showing that individuals, actions and consequences are situated in a wider context which has an impact along the forward path, at the same time as the individuals, actions and consequences are able to influence this wider context along the feedback path. The work of the Supplemental Ways for Implementing International Stability (SWIIS) Technical Committee involves the application of control strategies to non-technical problems, including conflict resolution. Conflict resolution issues are particularly important in Kosovo: a situation of significant ethnic conflict has existed for a considerable period of time and talks have recently started to try and resolve the status of the territory in such a way as to reduce the likelihood of further conflicts. Control theory can help model, and identify pathways through and beyond interethnic conflict. For example, at the most basic level, positive feedback loops indicate the potential for an exacerbation of problems and negative feedback mechanisms (calming) may be required in order to ensure a controlled and stable path through difficult negotiation periods. In Northern Ireland, another conflict zone, this was the case in the late 1990s. During this period, while important negotiations were underway, extremist nationalist elements in society planted a massive bomb in Omagh which killed and maimed scores of people and potentially destabilised a delicate negotiations process. Kosovan delegates at the panel session noted that in the current Kosovan negotiations there is a distinct lack of conflict management and control strategies which could be used to help to calm the situation in the context of an atrocity committed by extremist elements. It

114

was recognised that control strategies could be crucial in identifying solutions and potential problems before they appeared. Control strategies for a sustainable future need considerable attention and this should include consideration of open systems models and the identification of feedforward paths. The systems models will need to be highly dynamic and focus upon navigation paths rather than solely predictive and deterministic mechanisms. The systems involved are emergent and holistic in nature and futures methods might be useful. From the point of view of theoretical and engineering systems, SWIIS has focussed on the application of continuous time approaches which are well known from the field of process automation. These approaches have been particularly important in SWIIS work on sustainable futures. However, the fields of production automation and systems engineering have advanced over the past few years, giving rise to new methods which SWIIS could usefully apply. These new approaches include "Multi Agent Systems" (MAS). These systems have been very well known in software engineering for more than 20 years. More recently, a number of publications have been produced about their application in production automation. A multi agent system consists of a number of intelligent, co-operative and communicative hardware agents, such as robots, which are working on a common task. They can use their intelligence to divide the whole task into subtasks, as long as each subtask can be carried out by at least one of the agents. This procedure can be repeated to solve the common task. The most recent research is focussing on MMAS Multiple Multi Agent Systems, in which different multi agent systems are used to solve different aspects of a complex task. SWIIS has a number of similarities to a multi agent system. The actors are people with a distinct degree of intelligence and the ability to communicate and cooperate with others. A conflict could be defined as a competition between two or more multi agent systems, or as a multi multi agent system without co-operation between the different multi agent systems. This is very different from the case of production engineering, where the different multi agent systems co-operate with each other rather than acting counterproductively. Thus, the new approach of MMAS could provide a useful additional tool for the work of SWIIS.

115

This page intentionally left blank

AUTHOR’S INDEX Andreeski, C. ............................................... 55 Barsoum, N.................................................. 55 Bhattacharya, A. ......................................... 55 Carbone, G. ................................................ 31 Ceccarelli, M.......................................... 1, 31 Dimirovski, G.M.................................... 55, 71 Dinibütün , A.T. ...................................... 55,71 Erbe, H.H. ........................................... 21,113 Genser, R. ............................................ 49,113 Gojani, I. .................................................... 95 Hajrizi, E. ..................................................... 1 Han, M.W. .......................................... 107,113 Hersh , M.................................... 9, 61, 77,113 Icev, Z.A ...................................................... 71 Izworski , A............................................ 83, 89 Kieleczawa, A.. ............................................ 89 Kile, F.......................................................... 43 Kolemisevska, T. ................................... 55, 71 Kopacek, P. .......................... 1,21, 37, 95,113 Lewoc , J. B. .......................................... 83, 89 Ottaviano, E. .............................................. 31 Pajaziti, A. ................................................... 95 Schierer , E. .............................................. 101 Shala, G....................................................... 95 Skowronski , S ....................................... 83, 89 Stankovski, J. .............................................. 71 Stapleton, L........................................ 1,61,113 Vamos, T...................................................... 67 Vasant, P. ................................................... 55

117

This page intentionally left blank