基于改进粒子群算法的移动机器人路径规划  

Mobile Robot Path Planning Based on Improved Particle Swarm Optimization

作　　者：刘艳红[1] 陈田田 张方方 LIU Yanhong;CHEN Tiantian;ZHANG Fangfang(School of Electrical Engineering,Zhengzhou University,Zhengzhou 450001,China)

机构地区：[1]郑州大学电气工程学院

出　　处：《郑州大学学报（理学版）》2020年第1期114-119,共6页Journal of Zhengzhou University:Natural Science Edition

基　　金：国家自然科学基金项目(61473265,61803344,61773351,61603345);河南省引智计划项目(GZS2019008)

摘　　要：针对基本粒子群算法在路径规划时易陷入局部最优、规划路径较长等问题,提出了改进粒子群算法对移动机器人进行路径规划。首先使用MAKLINK图建立移动机器人的工作空间模型,然后采用Dijkstra算法搜索从起始位置到目标位置的全局次优无碰撞路径,最后将指数变量权重加入改进的粒子群算法中对次优路径进行优化,找到最短路径。与基本粒子群算法不同,改进粒子群算法中粒子不是向最优的粒子学习,而是向适应度值优于平均适应度值的粒子学习,并对低于平均适应度值的粒子进行变异处理。该方法能够提高粒子的多样性,避免粒子陷入局部最优。仿真结果验证了所提出的改进粒子群算法的有效性。An improved particle swarm optimization algorithm was proposed to tackle the local optimal and long planning path problem of the mobile robot.Firstly,MAKLINK graph was used to build the workspace model of mobile robot.Then,the Dijkstra algorithm was applied to seek a global sub-optimal collision-free path from the starting position to the target position.Finally,the improved particle swarm algorithm with an exponential variable weight was put forward to optimize the sub-optimal path and get the shortest path.Unlike the conventional algorithm,the particles in the improved algorithm were not learning from the optimal particle but from the particles whose fitness value were better than the average fitness value.The particles below the average fitness value were subjected to mutation processing.So the improved particle swarm optimization algorithm could improve the diversity of particles and avoid the local optimization.Simulation results illustrated the effectiveness of the proposed algorithm.

关 键 词：移动机器人 路径规划 MAKLINK图 DIJKSTRA算法 改进粒子群算法 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]