Task Offloading Based on Vehicular Edge Computing for Autonomous Platooning

Abstract

Autonomous platooning technology is regarded as one of the promising technologies for the future and the research is conducted actively. The autonomous platooning task generally requires highly complex computations so it is difficult to process only with the vehicle's processing units. To solve this problem, there are many studies on task offloading technique which transfers complex tasks to their neighboring vehicles or computation nodes. However, the existing task offloading techniques which mainly use learning-based algorithms are difficult to respond to the real-time changing road environment due to their complexity. They are also challenging to process computation tasks within 100 ms which is the time limit for driving safety. In this paper, we propose a novel offloading scheme that can support autonomous platooning tasks being processed within the limit and ensure driving safety. The proposed scheme can handle computation tasks by considering the communication bandwidth, delay, and amount of computation. We also conduct simulations in the highway environment to evaluate the existing scheme and the proposed scheme. The result shows that our proposed scheme improves the utilization of nearby computing nodes, and the offloading tasks can be processed within the time for driving safety. © 2023 CRL Publishing. All rights reserved.