摘要
采用链接图法建立了机器人工作空间模型 ;应用遗传算法规划多移动机器人运动路径 ;引入适应值调整矩阵新概念 ,以达到对多移动机器人运动路径的全局优化 ;基于面向对象技术 ,研制成功多移动机器人路径规划动态仿真系统 .大量仿真实验结果表明 ,所提方法可行 .
The paper presents a strategy for path planning of multi mobile robots using genetic algorithms (GAs). Obstacles in the 2D workspace of the multi mobile robots are simplified into convex polygons. The edges of the polygons are enlarged so that the size of the mobile robots can be scaled to points. MAKLINK Graph is used to model the workspace. Sequence numbers mark the cross points on the MAKLINK Graph. The sequence number is also used as a coding method for the chromosomes of GAs. Selecting randomly the cross points that connect the start point and the end one generates the initial population of the GAs. The length of the path is used as the corresponding fitness value of the chromosome. In order to avoid collision of any two mobile robots, a coordination matrix is put forward to decide the fitness values of the offspring. A simulation system is developed based on object oriented method. Simulation results show that the method is effective in path planning of multi mobile robots.
出处
《自动化学报》
EI
CSCD
北大核心
2000年第5期672-676,共5页
Acta Automatica Sinica
基金
国家教委优秀青年教师基金资助项目
关键词
路径规划
多机器人系统
遗传算法
优化
Path planning, multi mobile robots, genetic algorithms (GAs),optimization.
