Multi-UAV Collaborative Detection Based on Reinforcement Learning

In modern agriculture and industrial fields, UAV swarm has become an important research field. Because of the high complexity and uncertainty of UAV swarm task, the traditional control algorithm is difficult to deal with the requirements of practical application. The research work of this paper is the following: Firstly, the model of UAV detection is established. Then, Markov decision process model of multi-UAV is established, and state space, action space and reward function are designed. Secondly, use MADDPG algorithm to realize the multi-UAV area coverage and target tracking task. Finally, the proposed method is tested. The experimental results show that the performance of our multi-UAV cooperative detection algorithm is better than other algorithms.