A Study on Dynamically Reconfigurable Robotic System. 25th Report. Mechanism of Cooperative Behavior on Group Robotic System with Attractor.

Abstract

In this research, we address the organization of group behavior on decentralized autonomous robotic systems. Collective group behavior is exhibited in the natural world, such as by ants and fish, in teamwork in sports and by the human society. Therefore, research on group behavior of decentralized autonomous robotic systems can be regarded as one the research fields of Artificial Life. Decentralized autonomous robotic systems refer to multiple robotic systems including many autonomous robots, such as the Cellular Robotic System (CEBOT). The CEBOT, which has been studied by the authors, consists of a number of robotic units called cells. In research on the CEBOT, it is necessary to evolve a cooperative group behavior effectively in the system, since a well-organized group behavior is required to carry out given tasks efficiently and influences its performance ability. In order to organize the behavior in a dynamic environment, we proposed a concept of the self-recognition for the decision making of the behavior in a robotic group. In addition to the proposed concept, this paper will show a construction mechanism of group behavior using the character of the attractor. Based on this idea, we present the behavioral evolution of a group robotic system.