The autonomous evolution of cooperation activity of micro-robots by genetic algorithms(GAs)

摘要

Recently the studies concerning the multiple agents control system provoke strong interests in a lot of different research fields such as sensor system, vision procession, learning algorithm and so on. My interest is on whether and how to create a relatively common control mechanism to guide the design of such multiple agents system. In this paper a approach based on the generic algorithms(GAs) for the autonomous evolution of cooperative activities among agents is proposed. Here a micro-robots soccer-playing game such as that of MIROSOT (Micro-robot World Cup Soccer Tournament) is adopted as a standard example for the test of the approach.