%0 Journal Article
%T 一种无人机编队控制方法研究与仿真<br>Research and Simulation of a UAV Formation Control Method
%A 陈泽坤
%A 何杏宇
%J Modeling and Simulation
%P 2662-2672
%@ 2324-870X
%D 2024
%I Hans Publishing
%R 10.12677/mos.2024.133242
%X 现有的无人机群编队控制研究难以在障碍物环境下兼顾无人机的避障一致性和避障灵活性。为此，本文提出了一种基于强化学习的差异化无人机编队控制方法。该方法允许无人机群中的任一无人机根据其局部环境在编队聚集、编队保持和避障之间改变其编队控制策略，并设计了一个强化学习模型为上述三种控制策略下的无人机计算最优偏移向量。仿真结果表明，该方法可以有效地兼顾无人机群的避障灵活性和一致性，从而提高其飞行效率并保持其网络联通性。<br />
In existing formation control researches, it is hardly to reconcile both the obstacle avoidance consistency and obstacle avoidance flexibility of the UAV swarm. In view of this, this paper proposes a reinforcement learning based differentiated formation control method for UAVs. This method allows any UAV within the swarm to adapt its formation control strategy among formation aggregation, formation maintenance, and obstacle avoidance based on its local environment, and a reinforcement learning model is designed to compute the optimal offset vector for each UAV under the three aforementioned formation control strategies. Simulation results demonstrate that the method can effectively balance the maneuverability and consistency of obstacle avoidance for UAV swarms, thereby improving their flight efficiency and maintaining network connectivity.
%K 编队控制，避障一致性，避障灵活性，强化学习<br>Formation Control
%K Obstacle Avoidance Consistency
%K Obstacle Avoidance Flexibility
%K Reinforcement Learning
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=87410