关键词: 全局路径规划, 改进灰狼优化算法, 自适应变权重, 反向学习, 轮换捕食

Abstract: Aiming at the problems of the traditional robot path planning algorithm,such as low search efficiency and easy to fall into local optimal,a Multi-strategy improved Grey Wolf Optimization algorithm (MGWO) was proposed for path planning of mobile robots. Firstly,an adaptive variable weight strategy was proposed to improve the convergence speed by dynamically adjusting the weights;secondly,a reverse learning strategy was proposed to improve the global search ability;thirdly,a chain predation strategy was designed so that the search was guided by both the best individual and the previous individual; finally,a rotational predation strategy was proposed to improve the individual search ability of the algorithm. In order to verify the optimization performance of the MGWO algorithm,CEC2005 partial standard function was used to compare with the MGWO algorithm and the traditional grey wolf optimization algorithm,and the results showed that the optimization ability of MGWO algorithm was better than the traditional grey wolf optimization algorithm. MGWO algorithm was used to optimize the path of the mobile robot under raster maps in variety scales of 30×30,40×40 and 50×50 respectively,and the shortest path values obtained were 43.86,59.33 and 85.10,which were superior to improved genetic algorithm,sparrow search algorithm,improved artemisinin optimization algorithm and grey wolf optimization algorithm,thus verifying the feasibility of applying MGWO algorithm to path planning of mobile robots.

Key words: global path planning, improved grey wolf optimization algorithm, adaptive variable weight, reverse learning, alternate predation