火力与指挥控制2024,Vol.49Issue(2) :119-124.DOI:10.3969/j.issn.1002-0640.2024.02.018

融合Q学习算法和人工势场算法的无人机航迹规划方法

UAV Route Planning Method Based on Fusion of Q-learning Algorithm and Artificial Potential Field Algorithm

刘冬 余文泉 霍文健 李瑞 姜伟月
火力与指挥控制2024,Vol.49Issue(2) :119-124.DOI:10.3969/j.issn.1002-0640.2024.02.018

融合Q学习算法和人工势场算法的无人机航迹规划方法

UAV Route Planning Method Based on Fusion of Q-learning Algorithm and Artificial Potential Field Algorithm

刘冬 1余文泉 1霍文健 1李瑞 1姜伟月1
扫码查看

作者信息

  • 1. 北方自动控制技术研究所,太原 030006
  • 折叠

摘要

针对基于Q学习算法规划出的航线存在与静态障碍物发生碰撞危险的问题,提出融合Q学习算法和人工势场算法的航迹规划方法.该方法首先利用Q学习算法规划出一条航线,其次根据地图统计该航线每个航段内包含的障碍物,最后对每个包含障碍物的航段采用改进的人工势场法进行重新规划.实验结果显示,提出的融合方法能够在牺牲少量轨迹长度和时间的情况下,得到与静态障碍物避免发生碰撞的最短路径.

Abstract

Aiming at the problem of collision with static obstacles when using a flight route planned by Q-learning algorithm,a UAV route planning method based on fusion of Q-learning algorithm and artificial potential field algorithm is proposed.The method first uses the Q-learning algorithm to plan a route,then counts the obstacles contained in each flight leg of the route according to the map,and finally applies the improved artificial potential field method to re-plan the flight leg containing obstacles.The experimental results show that the proposed fusion method can plan the shortest route to avoid collision with static obstacles at the cost of a small amount of time and trajectory length.

关键词

航迹规划/Q学习算法/人工势场/无人机

Key words

route planning/Q-learning algorithm/artificial potential field/UAV

引用本文复制引用

出版年

2024
火力与指挥控制
火力与指挥控制研究会,火力与指挥控制专业情报网

火力与指挥控制

CSTPCDCSCD北大核心
影响因子:0.312
ISSN:1002-0640
参考文献量15
段落导航相关论文