基于双层微粒群优化的机器人全局路径规划  

Global robot path planning based on two-layer particle swarm optimization

作　　者：曾现峰[1,2] 张勇[1] 

机构地区：[1]中国矿业大学信息与电气工程学院,江苏徐州221116 [2]江苏联合职业技术学院徐州机电工程分院,江苏徐州221011

出　　处：《计算机工程与应用》2013年第19期238-241,共4页Computer Engineering and Applications

基　　金：国家自然科学基金(No.61005089);高等学校博士学科点专项科研基金资助课题(No.20100095120016)

摘　　要：采用微粒群优化解决机器人全局路径规划问题,近年来得到国内外学者广泛关注,并已经取得丰硕的研究成果。但是,已有成果往往难以应用于含有密集障碍物的环境。针对解决含有密集障碍物环境的机器人全局路径规划问题,提出一种双层微粒群优化方法。该方法通过底层微粒群优化,得到若干最优路径;通过顶层微粒群优化,在这些最优路径附近局部搜索,从而得到机器人的全局最优路径;通过对不可行路径实施脱障操作,使其成为可行路径。将所提方法应用于多场景的机器人路径规划,并与已有方法进行比较。实验结果表明,该方法能够找到机器人的全局最优路径。Solving the problem of global robot path planning using Particle Swarm Optimization（PSO）has attracted various researchers in recent years, and fruitful achievements have been obtained. Previous studies, however, are hard to be applied to environments containing dense obstacles. Aiming at solving the problem of global robot path planning in environments containing dense obstacles, a two-layer PSO is presented. In this method, several optimal paths are first obtained using the bottom layer PSO. The globally optimal path is then got by local search near these optimal paths using the up layer PSO. In addition, an infea-sible path becomes feasible by performing an escape obstacle operator. The proposed method is applied to solve the problem of robot path planning in various scenarios, and compared with previous methods. The experimental results confirm that the pro-posed method can find the globally optimal robot path.

关 键 词：机器人 路径规划 微粒群优化 脱障 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]