摘要
使用遗传算法求解多障碍离散路径规划问题时,容易产生大量无效解.通过计算个体的有效路径,评价个体,并在遗传操作中不断累积局部优势模式,可以对无效解进行遗传操作并最终生成有效解.无效解往往在有效路径的尾部陷入障碍.针对此变异操作,使得个体不仅可以保留前端累积的局部优势模式,同时可通过尾部变异跳出环境障碍.
It is very easy to generate a mass of invalid solution while solving path planning in environment containing large numbers of obstacles using genetic algorithm. By computing valid path to evaluate individuals and cumulate local advantage modules in the procedure of genetic operations, invalid individuals can be processed and formed valid individuals at last. Because invalid individuals are always barricaded in the end of its chromosomes, end-mutation is proposed to solve this problem. Simulation results show the validity of this method.
出处
《华侨大学学报（自然科学版）》
CAS
北大核心
2006年第3期317-320,共4页
Journal of Huaqiao University(Natural Science)
基金
福建省自然科学基金资助项目(A0540005)
关键词
路径规划
遗传算法
迷宫问题
变异算子
path planning, genetic algorithm, labyrinth, mutation operator
