Task Offloading in UAV-Assisted Vehicular Edge Computing Networks

As a promising architecture for supporting various intelligent vehicle applications, Vehicle Edge Computing (VEC) has received extensive research attention. However, during peak hours, the limited computing resources of VEC servers can make it difficult to meet the needs of delay-sensitive and computation-intensive tasks generated by a large number of vehicles. To overcome this challenge, we propose a UAV-assisted vehicle edge network that deploys a UAV equipped with mobile edge computing (MEC) capabilities as an aerial edge to alleviate the overload of VEC servers. To evaluate the performance of the network, the processing latency and energy consumption of tasks are incorporated into a system overhead construction. Moreover, we formulate a joint resource allocation and task offloading problem aimed at minimizing the system overhead. Since the formulated problem is proven to be NP-hard, we propose a hybrid algorithm based on genetic and simulated annealing algorithms (HGSAA), which can obtain a sub-optimal solution in polynomial time complexity. Simulation results demonstrate that HGSAA outperforms other benchmark schemes, achieving superior system performance. Simulation results show that HGSAA can achieve superior system performance compared to the other benchmark schemes.