摘要
为解决遗传算法(GA)在自主移动机器人路径规划中,过早收敛以及由于交叉和突变产生大量的不可行路径问题,对传统遗传算法进行了改进,采用二进制编码的方式来存储路径以便后续的交叉、变异等遗传操作。结合粒子群优化算法(PSO)进行局部搜索,加快了遗传算法的搜索速度,提高了搜索效率。同时引入修复机制,通过利用修复机制研究所有的不可行路径,并确定其不可行的原因进行修正。仿真结果表明,在单目标简单情况下,改进的遗传算法具有更快的收敛速度同时避免了局部最优,在多目标复杂环境下,能够得到合适的路径解。
In order to solve the premature convergence of genetic algorithm(GA) in the path planning of autonomous mobile robots and a large number of unfeasible paths due to crossover and mutation, the traditional genetic algorithm is improved. The binary coding was adopted to store the paths for subsequent crossover, mutation and other genetic operations. Combined with particle swarm optimization(PSO), local search was carried out to speed up the search speed of genetic algorithm and improve the search efficiency. The repair mechanism was introduced to study all the infeasible paths, and determine the infeasible reasons for correction. The simulation results show that the improved algorithm has faster convergence speed and avoids local optimization in the case of simple single target. In the multi-objective complex environment, the appropriate path solution can be obtained.
作者
陈高远
宋云雪
Chen Gaoyuan;Song Yunxue(School of Aeronautical Engineering,Civil Aviation University of China,Tianjin 300300,China)
出处
《计算机应用与软件》
北大核心
2023年第2期302-307,共6页
Computer Applications and Software
关键词
遗传算法
路径规划
修复机制
粒子群优化算法
二进制编码
Genetic algorithm
Path planning
Repair mechanism
Particle swarm optimization
Binary coding
