李靖 杨帆 韩艳芬 陈虹

摘  要： 人工势场法对机器人进行路径规划时，存在障碍物使目标不可达、死锁和局部极小点等问题，容易导致路径规划失败，此外发现即使路径规划成功，由于斥力的存在，必定使得规划路径不是最短、最优的。针对上述问题提出一种动态引力场路径规划算法，该算法完全去掉斥力场，将引力场的大小及方向动态化，虚拟目标点动态化，不断交替寻找最优或次优方向进行避障，动态地修正路径，最后对路径进行平滑优化，规划出最短、最优的路径。该方法解决了人工势场法固有的缺陷，同时优化了规划路径的长度，仿真实验结果说明该算法具有实用性和有效性。

关键词： 机器人路径规划; 动态引力场; 动态路径修正; 路径长度优化; 路径平滑优化; 路径优化对比

中图分类号： TN99?34; TP242                     文献标识码： A                     文章编号： 1004?373X（2020）11?0041?06

Research on robotic path planning based on dynamic gravitational field

LI Jing1， YANG Fan1， HAN Yanfen2， CHEN Hong2

（1. School of Electronic & Information Engineering， Hebei University of Technology， Tianjin 300401， China;

2. Tianjin Light Industry Vocational Technical College， Tianjin 300350， China）

Abstract： Some problems may occur when the artificial potential field method is applied to robot path planning， for example， target is unreachable due to obstacles， deadlock and local minimum point may occur， which can easily lead to failure of path planning. In addition， it is found that even if the path planning is successful， the planned path is definitely not the shortest and optimal due to the existence of repulsive force. In view of the above， a path planning algorithm based on dynamic gravitational field is proposed. In the algorithm， the repulsive force field is completely removed， both the size and direction of gravitational field and the virtual target point are changed dynamically to continuously search the optimal or suboptimal directions alternately for the obstacle avoidance， so that the path is corrected dynamically. In the end， the path is subjected to evenness optimization and the shortest and optimal path is planned. With this method， the inherent defects of the artificial potential field method are removed， and the length of the planning path is optimized. The simulation results show that the algorithm is practical and effective.

Keywords： robotic path planning; dynamic gravitational field; dynamic path correction; path length optimization; path evenness optimization; path optimization contrast

0  引  言

随着机器人技术飞速发展，路径规划或障碍物规避问题成为机器人调度策略等控制领域研究的重点和热点问题之一，是机器人顺利完成各项作业任务的前提条件。机器人路径规划技术是根据规划空间中障碍物信息，在满足一定约束条件和评价指标的条件下，线下或线上为机器人规划出从出发点到目标点合理避开障碍的满意路径。

机器人路径规划根据环境信息掌握情况[1]，可以分为全局规划和局部规划搜索算法。全局规划算法主要应用于障碍物地图信息已知的情况，启发式算法运用居多，如遺传算法[2]、粒子群算法[3]等。……