自动化学报2025，Vol.51Issue(5)：905-916,12.DOI:10.16383/j.aas.c240519
性能函数引导的无人机集群深度强化学习控制方法
Performance Function-guided Deep Reinforcement Learning Control for UAV Swarm
摘要
Abstract
A novel performance function-guided deep reinforcement learning control method is proposed for the un-manned aerial vehicle(UAV)swarm system,which simultaneously evaluates both the demonstration experience from the performance function and exploratory actions from the learning strategy to guarantee efficient and reliable policy updating,achieving high-performance control of the UAV swarm system.Firstly,based on the leader-follow-er framework,the UAV swarm control problem is transformed into a tracking problem under the leader-follower paradigm,and then,the model-based tracking control is proposed,where the performance function is designed to constrain the tracking error within a given range,thereby achieving UAV model-driven formation control.Then,to address the invalid problem of performance function under complex working conditions,the deep reinforcement learning and the performance function-driven methods are combined to propose the performance-function-guided deep reinforcement learning control method,where the demonstration of performance function is used to assist in training the reinforcement learning network.By jointly evaluating exploratory and demonstrative actions,the pro-posed method ensures a learned policy that significantly outperforms the performance function-driven control alone,effectively enhancing the accuracy and robustness of UAV formation control.Comparative experimental results show that the proposed method significantly improves the control performance of UAV swarms,realizing high-per-formance swarm control with both robustness and flight accuracy.关键词
无人机集群/深度强化学习/引导式学习/智能编队控制Key words
Unmanned aerial vehicle swarm/deep reinforcement learning/guided learning/intelligent formation control引用本文复制引用
王耀南,华和安,张辉,钟杭,樊叶心,梁鸿涛,常浩,方勇纯..性能函数引导的无人机集群深度强化学习控制方法[J].自动化学报,2025,51(5):905-916,12.基金项目
科技创新2030"新一代人工智能"重大项目(2021ZD0114503),国家自然科学基金(62403190,62427813,62433010)资助Supported by the National Key Research and Development Program of China(2021ZD0114503)and National Natural Sci-ence Foundation of China(62403190,62427813,62433010) （2021ZD0114503）
