无人机集群路径规划算法研究综述

A summary of UAV swarm path planning algorithm research

无人机集群路径规划算法是无人机集群控制的重要研究方向之一。多无人机路径规划相较于单无人机路径规划,会考虑空间协同和时间协同约束、飞行安全等问题。首先对无人机集群路径规划算法进行分类,分为传统路径规划算法、智能优化算法和深度强化学习算法。其次对各类算法进行简要分析和总结,针对缺陷,给出相应的改进思路及例子。再着重对深度强化学习算法在无人机集群路径规划上的应用进行讨论。最后对无人机集群路径规划算法进行总结,对下一步的改进与研究方向进行展望。

UAV swarm path planning algorithm is one of the important research directions of UAV swarm control.Compared with single UAV path planning,multi-UAV path planning considers space coordination and time coordination constraints,flight safety and other issues.The UAV swarm path planning algorithms are classi⁃fied firstly,which is divided into traditional path planning algorithm,intelligent optimization algorithm and deep reinforcement learning algorithm.Secondly,the various algorithms are briefly analyzed and summarized,and cor⁃responding improvement ideas and examples are given for the defects.Then the analysis of the application of deep reinforcement learning algorithm in UAV swarm path planning is discussed emphatically.Finally,the UAV swarm path planning algorithm is summarized and the next improvement and research direction are prospected.