Deep Reinforcement Learning Based Collaborative Optimization of Communication Resource and Route for UAV Cluster

To solve the communication service quality reduction and real-time route planning difficulty issues caused by inter-node interference in typical logistics environments, this paper proposes a deep reinforcement learning based method to realize the collaborative optimization of communication resource allocation and route planning for the UAV cluster. Simulation experiments show that the communication agents among UAV nodes can effectively learn to minimize the communication transmission interference while ensuring the latency constraint. The system communication capacity obtained by the proposed approach is about 2.15 times of the random resource allocation method. Meanwhile, a reasonable route is planned for each UAV node to facilitate the completion rate of logistics distribution tasks to reach 81.25%.