Evolving Mobile Robots Able to Display Collective Behaviours
Baldassarre Gianluca, Nolfi Stefano, Parisi Domenico

Abstract:

In this paper we present a set of experiments in which a group of simulated robots were evolved for the ability to aggregate and to move together toward a light target. Evolved individuals display interesting behavioural patterns in which groups of robots act as a single unit. Moreover, groups of identical evolved individuals display primitive forms of "situated" specializations in which different individuals play different behavioural functions according to the circumstances. Overall, the results presented in the paper demonstrates that the evolutionary method, by allowing the synthesis of behaviours that emerge from the interaction between the robots and the environment and between different individual robots, is a powerful method for synthesizing collective behaviour.