A Research On Multi-Robot Cooperation Mechanism Based On PSO Optimizer

With flexible structure and high adaptability, multi-robot system (MRS) can be widely used in many areas such as astronautics, national defense, industry, service and so on. To take the advantage of MRS, the reasonable design of cooperation mechanism and system optimization using algorithms are the key of the problem. Therefore, this paper takes three classic cooperation problems (hunting, foraging, and observation of multiple targets) as test platform, and uses modified PSO optimizer to build optimized group behaviours of MRS for improvement of the cooperation efficiency and system performance.This paper proposes the ideas of neighborhood operator, cluster analysis, adaptive velocity factor to solve different problems based on the basic concept of PSO. The neighborhood operator of PSO is proposed for the multi-robot hunting to build the ring formation for collaborative search, and its performance is analyzed. The searching method of global patrol and local exploit is achieved as well as task-oriented work-division module. The relationship among the collaborative search in first period, the performance of the transportation and the whole process of foraging is analyzed. The adaptive PSO is designed for Multi-Robot Observation of Multiple Moving Targets problem based on group division of multi-target using cluster module. And its performance is compared with the traditional CMOMMT and A-CMOMMT methods; moreover, it solves the problem of local optimization. Simulation shows that using PSO to optimize MRS can highly improve the cooperation efficiency and system performance.