摘要
针对动态环境中多移动机器人路径规划问题,提出了一种基于知识的遗传算法。该算法路径表示简单,所设计的基于知识的遗传算子并入了领域知识和局部搜索技术,同时,把领域知识并入初始种群,大大提高了遗传算法的效率;将协调路径适应值矩阵引入基于知识的遗传算法,实现了多移动机器人协调路径规划。仿真结果验证了该算法在多移动机器人路径规划中的可行性和有效性。
For path planning of the multi- mobile robot in the dynamic environment, the authors propose a knowledge -based genetic algorithm (GA). The algorithm has simple path representation, six designed genetic operators with domain knowledge and a local search technique. The proposed knowledge based genetic algorithm incorporates the domain knowledge into its initial population, which improves efficiency of the algorithm. In order to plan path of any two mobile robots in the dynamic environment, a coordination matrix is put forward to decide the fitness values of the offspring. The feasibleness and effectiveness of the proposed genetic algorithm in path planning for multi - mobile robot is demonstrated by simulation.
出处
《黑龙江大学自然科学学报》
CAS
北大核心
2006年第2期195-198,201,共5页
Journal of Natural Science of Heilongjiang University
基金
国家自然科学基金资助项目(69985002)
国家863高科技发展计划资助项目(2001AA422270)
关键词
多移动机器人
路径规划
遗传算法
multi - mobile robot
path planning
genetic algorithm
作者简介
周兰凤(1966-),女,讲师,硕士,博士研究生,主要研究方向:智能机器人,智能控制,自主导航
通讯简介:洪炳熔(1937-),男(朝鲜族),教授,博士,博士生导师