Cooperative Hunting Of Multiple Mobile Robots Under Unknown Dynamic Environment

Single robot can not fulfill the demands of one complicated task sometimes, so the alternative approach, multi-robot cooperation, turns to be more and more significant. Nowadays, the cooperation robotics, as a rising cross subject, integrates the theories of distributed AI, sociology, management science and biology etc. It discusses many topics systematically, such as architecture, communication, cooperative behaviors and evolution of the robot system. Especially, multi-robot cooperative control guarantees the successes of large object transportation, space exploration, search-rescue and some other tasks. To satisfy these requirements, researches on multi-robot cooperative control has become a focus of robotics.Firstly, the paper surveys the developments of cooperative robotics and several influential multiple robots systems. Following the main topics about multiple mobile robots, the main research methods, the characteristics and key technology of cooperative control of multiple robots are introduced. The paper introduces the sensor models which describe the information perception method about the environment. Also this paper explained the working principle of LRF.For obstacle avoidance behavior, this paper adopts the method of fuzzy logic control. The collision risk based obstacle avoidance scheme is used to keep away from random obstacles during hunting. The synthesized behavior is obtained by fusing avoidance behavior, encirclement behavior and pincer attack behavior.To realize successful hunting, this paper studies the formation, point stabilization, "potential points" and the fuzzy genetic algorithm and so on. Through these strategies, hunter robots continually approach to the target and adjust their motions opportunely. In order to improve the efficiency of encirclement, terminal processing algorithm of circular motion is used to encircle the target. The synthesized behavior is obtained by fusing avoidance behavior, encirclement behavior and pincer attack behavior.Hunter robots'motion is constantly checked for the adaptive ability and fault tolerance. The failure of the hutting robots will be eliminated by the hunting program. According to the number of robot who is in good condition decide to adopt. A hunting strategy will be chosen according to the number of the robot who is in good condition. Simulated hunting experiments were conduct under different conditions in Microsoft Robotics Studio (MRS) and simulation results demonstrate the correctness and effectiveness of the proposed scheme.