Evolution of Signalling in a Group of Robots Controlled by Dynamic Neural Networks
Ampatzis Christos, Tuci Elio, Trianni Vito, Dorigo Marco

Abstract:

Communication is a point of central importance in swarms of robots. This paper describes a set of simulations in which artificial evolution is used as a means to engineer robot neuro-controllers capable of guiding groups of robots in a categorisation task by producing appropriate actions. In spite of the absence of explicit selective pressures (coded into the fitness function) which favour signalling over non-signalling groups, communicative behaviour emerges. Post-evaluation analyses illustrate the adaptive function of the evolved signals and show that they are tightly linked to the behavioural repertoire of the agents.