Around ten years ago, when we were both PhD students, working on different but related aspects of artificial intelligence, we shared an office in the furthest corner of the Department of Computer Science at University College London. Our friendship began then, but our professional collaboration only really got going when we both left, one of us moving the few yards to the University of Westminster and the other further afield to the University of Warwick and later the University of Southampton. Nevertheless, we can trace back many of our inspirations to those days at UCL, in discussions with Derek Long, John Campbell, Maria Fox and John Wolstencroft, who all contributed to our initial enthusiasm for working in this area. On leaving UCL, however, we tried to bring our research interests together in the newly emerging area of agent-based systems, but found difficulties in communica tion with each other over basic terms and concepts, simply due to the immaturity of the field. In other words, the problems we had in finding a base on which to develop our ideas set us on a long path, over a number of years, resulting in our construction and refinement of a conceptual framework within which to define, analyse and ex plore different aspects of agents and multi-agents systems. This is the work reported in this book. This book presents a formal approach to dealing with agents and agent systems. The Z specification language is used to establish an accessible and unified formal account of agent systems and inter-agent relationships. In particular, the framework provides precise and unambiguous meanings for common concepts and terms for agent systems, allows for the description of alternative agent models and architectures, and serves as a foundation for subsequent development of increasingly refined agent concepts. The practicability of this approach is verified by applying the formal framework to three detailed case studies. The methodology presented takes a very significant step towards organising and structuring the diverse and disparate landscape of agent-based systems by applying formal methods to develop a defining and encompassing agent framework. The book will appeal equally to researchers, students, and professionals in industry Mark d'Inverno and Michael Luck present a formal approach to dealing with agents and agent systems in this second edition of Understanding Agent Systems. The Z specification language is used to establish an accessible and unified formal account of agent systems and inter-agent relationships. In particular, the framework provides precise and unambiguous meanings for common concepts and terms for agent systems, allows for the description of alternative agent models and architectures, and serves as a foundation for subsequent development of increasingly refined agent concepts. The practicability of this approach is verified by applying the formal framework to three detailed case studies. The book will appeal equally to researchers, students, and professionals in industry. Front Matter....Pages I-XVII The Agent Landscape....Pages 1-13 The smart Agent Framework....Pages 15-38 Agent Relationships....Pages 39-67 An Operational Analysis of Agent Relationships....Pages 69-90 Sociological Agents....Pages 91-123 The Contract Net as a Goal Directed System....Pages 125-139 Computational Architecture for BDI Agents....Pages 141-151 Evaluating Social Dependence Networks....Pages 153-167 Conclusions....Pages 169-172 Back Matter....Pages 173-191
