基于改进粒子群算法的机器人路径规划  

Robot path planning based on improved particle swarm algorithm

作　　者：王志中 WANG Zhizhong(Chongqing Jianzhu College, Chongqing 400072, CH)

机构地区：[1]重庆建筑工程职业学院,重庆400072

出　　处：《制造技术与机床》2018年第2期150-154,共5页Manufacturing Technology & Machine Tool

基　　金：国家自然科学基金资助项目(60974138)

摘　　要：为了提高移动机器人路径规划的质量,提出了基于改进粒子群算法的机器人路径规划方法。对障碍物进行膨化处理,简化了障碍物模型;通过坐标变换,将二维优化问题简化为一维优化问题;建立了包含路径长度和路径平滑度的适应度函数;分析了传统粒子群算法及缺陷,引入了跳出机制和牵引操作,跳出机制保持了种群多样性和全局搜索能力,牵引操作加快了算法收敛速度,从而提出了改进粒子群算法;经仿真实验验证,改进算法规划的路径在长度、平滑度、规划时间上均具有优势。To improve path planning quantity of mobile robot, path planning method based on improved particle swarm algorithm is proposed. Obstacles are expanded to simplify their model, and two-dimensional opti-mization is simplified to one-dimensional optimization. Fitting function is built including path length and evenness. Principle and shortcomings of traditional particle swarm algorithm are analyzed, and jumping mechanism and pulling operation are introduced. Jumping mechanism can keep population diversity and global searching ability, and jumping operation can run-up convergence speed of algorithm, so that improved particle swarm algorithm is put forward. By simulation experiment, improved algorithm possesses advantages of path length, evenness and planning time.

关 键 词：移动机器人 路径规划 改进粒子群算法 跳出机制 牵引操作 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]