Unmanned Aerial Vehicles (UAV)-assisted vehicular network is an effective approach to improve quality of service for vehicular applications. However, achieving a balance between system delay and energy consumption remains a long-standing challenge. In this paper, task offloading, resource allocation and trajectory are jointly optimized. We employ block coordinate descent to decompose the problem into three sub-problems with lower computational complexity. Specifically, in the stage of optimizing UAV trajectories, we design a heuristic algorithm to select positions at various granularity levels. For task offloading, a hyper-heuristic algorithm is proposed. We utilize tabu search to augment the improved genetic algorithm, compensating for its disadvantage of converging to local optima. For computing resource allocation, the Lagrangian multiplier method is used to solve the sub-problem. The simulation results demonstrate that the proposed algorithm can reduce system costs by 21.39% compared to the benchmark algorithms.