Resource Allocation in Vehicular Networks with Multi-UAV Served Edge Computing

With the rapid development of intelligent transportation systems, there is an increasingly strong demand for low-latency and high-bandwidth vehicular services, such as automatic driving assistance, emergency alarm, and infotainment. However, in some cases (e.g., traffic congestion, remote areas), the ground communication networks alone cannot meet the vast needs of vehicles. Unmanned aerial vehicles (UAVs) are flexible and deployable, which can be used as a supplement to the ground networks, to relieve the communication pressure on ground facilities, such as base stations. In this paper, we use multiple UAVs to provide services for vehicles and model the multi-UAV scenario as a collaborative multi-agent system. All UAVs share limited bandwidth resources and equip with edge computing servers to serve the vehicles. In addition, serious consequences may be caused if the delay requirements of vehicles are not satisfied. Therefore, we take vehicle safety as the top priority and the delay requirement as the constraints. Then we exploit the Lagrange multiplier to combine the constraint function and cost function, so as to reduce the resource consumption as much as possible on the premise of ensuring the safety of the vehicles. The influence of channel efficiency and computing power should also be taken into account when allocating resources. We adopt the multi-agent reinforcement learning to train the UAVs, and meanwhile introduce the attention mechanism so that each UAV can optimize itself better with the information of other UAVs. Through a large number of experiments, the effectiveness of our proposed method is verified. Particularly, in the case of strictly limiting bandwidth resources, resources can still be allocated according to vehicle needs under the premise of ensuring vehicle safety.