Coordinated Control Technology Of Multiple Mobile Robots Based On Multi-Agent

The problem of multi-robot cooperation and coordination is central to mobile robotics. Cooperation and coordination will improve the intelligent performance of robots and can complete lots of impossible missions for single robot. The research on multi-robot cooperation and coordination is of great academic and applied significance.Based on analyzing the development prospect and current research of the multi-robot system in depth, this thesis demonstrates that it's reasonable and effective to make full use of research results in present artificial intelligence field, especially the theory of MAS (Multi-agent system) in the distributed artificial intelligence field for the research of the multi-robot system. In terms of this opinion, the thesis carries on the research of the mufti-robot system based on the theory of MAS. This thesis is supported by the Hi-tech Research and Development Program of China (No. 2003AA430030). The detailed content in the thesis is described in follows:1.Based on the further investigation of the MAS theory, put forward a coordinated control of Multi-robot system model based on Multi-agent technology. It integrates effectively decision of tasks, planning, information and real-time execution. It can also unify the function of system and robots.2. Based on the WLAN (wireless local area network), established the multi-robot communication system. The WINSOCK control is used in software realize the server/client full duplex relation between robot and control centre.3. The thesis analyses further the structure of multiple mobile robots formation system, put forward a kind of multiple mobile robots formation control. It planned the robot team based on maintain formation, and the movement course of robot team is divided into normal movement phase, obstacle avoidance phase and formation recovery phase. At the same time, fuzzy control is applied to avoid barriers independently.Finally, the control system is simulated in the MOBOTSIM and the experiment's results verify that the above strategy and algorithm is correctness and effectiveness.