Evolutionary Methods in Self-organizing System Design

Self-organizing systems could serve as a solution for many technical problems where properties like robustness, scalability, and adaptability are required. However, despite all these advantages and due to the decentralized control there is no straight-forward way to design such a system. In this paper we describe a novel design approach using genetic algorithms and artificial neural networks to automatize the part of the design process that requires most of the time. A simulated robot soccer game was implemented to test and evaluate the proposed method. A new approach in evolving competitive behavior is also introduced using Swiss System instead of the full tournament to cut down the number of necessary simulations.