Trajectory Planning and Energy Optimization for UAV-Assisted Vehicular Networks in Urban Scenarios

Unmanned aerial vehicle (UAV) assisted edge networks can provide convenient, fast, and flexible services for ground users (GEs) through line-of-sight (LoS) communication, and the Internet of vehicles (IoV) needs to upload and update the information of vehicles effectively to guarantee the safety of road traffic. Therefore, UAV-assisted vehicular networks play a crucial role in some special circumstances. Due to the occlusion of buildings, the communication between UAVs and vehicles in urban scenarios needs to comprehensively consider the LoS communication and non-line of sight (NLoS) communication. In this paper, we investigate the trajectory planning of the UAVs in UAV-assisted vehicular networks to ensure the effective uploading of vehicles' information and minimize the energy consumption of the UAVs. The objective is the cumulative service time of multiple UAVs, considering the continuous movement action and discrete service selection decision. We propose a multi-agent deep deterministic policy gradient (MADDPG) algorithm to intelligently tackle the optimization problem. The simulation results demonstrate that the proposed algorithm can effectively guarantee the uploading process of vehicles' information through optimizing the trajectory of UAVs. Moreover, the results verify that the cooperation of multiple UAVs can reduce the completion time of tasks, and the overall energy consumption of the UAVs is lower than that of a single UAV.