A Task Offloading Method Based on Genetic Algorithm in Vehicular Networks

As intelligent transportation systems and smart connected vehicles continue to evolve, the issue of task offloading has garnered extensive attention. Vehicle Fog Computing (VFC) represents an emerging computing paradigm that capitalizes on the unused storage and computing resources of vehicles to offer services to other users. This approach helps mitigate the challenges associated with high communication latency and bandwidth constraints in task offloading scenarios.This paper proposes an optimization method based on the Genetic Algorithm (GA) to address the task offloading problem in VFC environments, aiming to minimize task processing latency. By constructing a joint optimization framework that comprehensively considers the task computation requirements, data volume, maximum tolerable delay, and resource constraints of vehicles and fog nodes, an efficient GA model is designed. The global search capability of GA is utilized to achieve the collaborative optimization of task offloading decisions and resource allocation. Simulation results show that compared with greedy algorithms, random algorithms, and local processing strategies, the proposed method can significantly reduce system latency under different numbers of service vehicles, demonstrating superior scheduling efficiency and stability.