Evolving Collective Control, Cooperation, and Distributed Cognition

Abstract

Studies in collective robotics usually emphasize aspects likeefficiency, robustness, and flexibility of the system. These areall desirable features for a robotic system, which, however, donot come for free along with the distributed approach. Suitabledesign methodologies must be devised to obtain similar featuresin a collective robotic system for what concerns both the robotichardware and the control algorithms. In this chapter, we discuss theproblem of synthesizing distributed controllers for a group of robotsby using evolutionary techniques.