基于改进遗传算法和改进人工势场法的复杂环境下移动机器人路径规划

The Path Planning of Mobile Robots based on the Improved Genetic Algorithm and the Improved Artificial Potential Field Algorithm in Complex Environment

为了提高移动机器人在复杂环境下的路径规划能力,通过双层路径规划思想研究了移动机器人路径规划问题。用栅格法对机器人工作环境进行建模,首先采用改进的遗传算法进行全局路径规划,解决了由于交叉概率和变异概率选择不当导致最优个体丢失的问题;然后在规划好的全局路径的基础上利用改进的人工势场法进行局部动态避障,解决了局部极小点问题。结果表明:静态环境下,采用改进遗传算法规划出的最优路径,与传统遗传算法相比其长度缩短了1. 47 m,收敛速度加快;动态环境下,采用改进人工势场法进行路径规划,所用时间与基本人工势场法相比缩短了7. 24 s;复杂环境下,移动机器人采用双层路径规划思想能够规划出一条优化路径。可见改进后的算法是有效的。

In order to improve the path planning ability of mobile robots in complex environment,a two-level path planning idea was used to investigate the path planning of mobile robots.The mobile robots working environment was modeled with the grid method.First,the improved genetic algorithm was used to plan the global path,and optimal individual loss problem caused by the improper selection of crossover probability and mutation probability was solved;Then,the improved artificial potential field algorithm was used to avoid the local dynamic obstacles based on global path,and the problem of local minima was solved.The results show that in static environment,compared with the traditional genetic algorithm,the length of the optimal path planned by improved genetic algorithm is shortened by 1.47 m and the speed of convergence is improved;in dynamic environment,compared with the traditional artificial potential field algorithm,the time of using the improved artificial potential algorithm to plan path is reduced by 7.24 s and the mobile robot can plan an optimization path by using a two-level path planning method in complex environment.It can be seen that the improved algorithm is effective.