摘要
将遗传算法用于路径规划时,传统算法虽然简单,但不适用转弯情况较多的复杂地图。针对这一问题,首先将RRT算法用于栅格环境下产生初始路径,其次提出一种新的插入算子,最后进行路径优化。根据不同地图与其他文献中的改进遗传算法,进行对比研究与分析,制定路径长度与算法用时2个指标来评判算法的优劣。仿真结果表明,改进算法得到的路径长度缩短了70%,路径长度达到最优的用时减少了8%。
The genetic algorithm is not suitable for path planning of complex terrain maps with many turns,although it is simple. In view of this problem,RRT algorithm is used to generate initial path in grid environment,then a new insertion operator is proposed,and finally path optimization is carried out. According to the improved genetic algorithm in different maps and other literature,comparative study and analysis are carried out. Two indexes,path length and algorithm time,are established to evaluate the algorithm. The simulation results show that the path length obtained by the improved algorithm is reduced by 70%,and the time required to reach the optimal path length is reduced by 8%.
作者
宋宇
王志明
SONG Yu;WANG Zhiming(School of Computer Science and Engineering,Changchun University of Technology,Changchun 130012,China)
出处
《现代电子技术》
北大核心
2019年第24期172-175,共4页
Modern Electronics Technique
基金
吉林省青年科研基金资助项目（20160520020JH）
吉林省“十三五”科学技术研究资助项目（2016第342号）~~
关键词
移动机器人
遗传算法
路径规划
算法评判
插入算子
路径优化
mobile robot
genetic algorithm
path planning
algorithm judgment
insertion operator
path optimization
