机器人全覆盖最优路径规划的改进遗传算法  

Improved Genetic Algorithms optimal area covering path planning for family robot

作　　者：刘松[1] 李志蜀[1] 李奇[2] 

机构地区：[1]四川大学计算机学院,成都610064 [2]四川师范大学软件重点实验室,成都610066

出　　处：《计算机工程与应用》2009年第31期245-248,共4页Computer Engineering and Applications

基　　金：科技部科技型中小企业创新基金(No.06CC26225101730);四川省科技攻关计划No.05GG021-003-2~~

摘　　要：全区域覆盖是一种特殊的路径规划,要求遍历环境中所有的可达区域。目前已经提的许多算法,如模板算法、分块算法等,都只能保证覆盖所有的区域,对于寻找全局最优解却无能为力。提出了一种基于遗传算法的全区域覆盖算法,结合分块算法和模板算法的优点。先采用矩形分解法将环境划分成若干个相邻的子模块,并为每一个子模块选用相应的模板,从而生成覆盖路径,然后采用遗传算法找出最优的路径。算法在虚拟环境中进行了实验,实验结果证明了其可行性和有效性。A special kind of path planning is complete coverage path planning.There are a lot of algorithms on this problem have been developed,e.g.template based,cellular decomposition.But these algorithms just cover the complete area;they are not designed to optimize the process.This paper presents a method of complete coverage path planning based on genetic algorithms, which combine the advantages of cellular decomposition and template algorithm.The environment is divided in sub-regions as in rectangular decomposition method,and then Genetic Algorithms （GA） is used to compute and find the order of the sub-regions and the appropriate template for each region. The algorithm is tested in the virtual environment;the simulation results confirm the feasibility of this method.

关 键 词：全区域覆盖路径规划 遗传算法 矩形分解法 模板算法 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]