Complexity science for the design of swarm robot control systems

Abstract

Swarm robots and their neighbourhood structures are defined and illustrated. The robots interact through these neighbourhood structures. Elements of complexity science can be used in the design and control of swarm robot systems. These include the emergent behaviour from the local interaction of agents, chaotic dynamics, relational structures for representation, learning and co-evolution, and simulation science. The concepts are presented using an example of football playing robots.