Joint task offloading and resource allocation for vehicle edge computing

Abstract As autonomous driving and in-vehicle applications develop rapidly, vehicles face significant computing challenges. In order to address this issue, vehicle edge computing (VEC) has emerged. In this paper, we study the issue of time delay and energy consumption for task offloading in VEC systems. Considering varying computing resources and changes in vehicle position due to vehicle mobility, we integrate offloading decisions with resource allocation. The problem is summarized as minimizing the delay and energy consumption of completing vehicle tasks. To solve the multi-objective optimization solution, we propose an improved multi-objective particle swarm optimization algorithm (IMOPSO). At first, we initialize particles by entropy weight method and Roulette Wheel Selection. Then, we introduce the transition probability to control the particles’ exploration in the search space for optimal solutions. Extensive simulation results demonstrate that the IMOPSO algorithm is superior to other alternative algorithms.