Cooperative control. A post-workshop volume. 2003 Block Island workshop on cooperative control, Block Island, RI, USA, June 9--11, 2003. (Q1768939)

[The articles of this volume will not be indexed individually.] This book presents some publications from a workshop on cooperative control held on Block Island (Rhode Island) in June 2003. The fifteen papers deal mainly with applied aspects of cooperative systems with emphasis on the mobile robotics area. For two similar volumes, see [Cooperative control and optimization. Boston: Kluwer (2002; Zbl 0999.93005)] and [Cooperative control: models, applications and algorithms. Dordrecht: Kluwer (2003; Zbl 1013.93002)]. Here, most articles deal with cooperation issues for mobile vehicles which are operating in swarms and have to reach certain goals. The inspiration is sometimes biological, like schools of fish and swarms of birds. A key concept is the one of decentralized observation and control (attraction or repulsion from neighbours). The goal is often to stabilize an associated dynamics. But tracking problems are involved too (pursuit problems). Issues of collision avoidance and rendez-vous problems are addressed. The methods vary: predictive control, integer programming, graph theory methods, optimization. The authors propose certain perspectives which seem to be relevant for this context: the sharing of information with graph-theoretic consequences, relating the geometry of the formation to sensing, relating the synchronization of oscillators to the stabilization of collective motion. Switching is proposed as a method to circumvent certain obstacles. Real-time motion is considered, too. Two publications are somewhat distinct: One deals with laboratory observation of the behavior of schooling fishes. Another, quite original one (and originality is needed for this kind of subject which is not understood), of F. Bullo and J. Cortés, deals with gradient flows in a discontinuous context. Overall, the subject remains mysterious and this is confirmed by some sentences from the text: ``It is perhaps too early to be able to list the fundamental problems arising in this problem area'' (p. 119), so that if problems are not clear, solutions are even less. The authors mainly try to adapt the issues involved to their own methods, or limit themselves to ``descriptive arguments and simulation examples'' (p. 137).