Interruption-Aware Task Offloading for Vehicular Edge Computing of Parking Lots

Abstract

Vehicular Edge Computing (VEC) provides reliable and efficient offloading services for increasingly latency-sensitive computing tasks in the Internet of Vehicles (IoV). As an important paradigm of VEC, parking vehicles (PVs) with idle computing resources can effectively accomplish offloading tasks. However, the heterogeneity of the computing resources among various PVs and the dynamic nature of their parking states lead to significant challenges to the design of offloading mechanisms. To address the above-mentioned issues, this paper formulates an optimization problem that considers service interruption risk and aims to minimize the average delay while adhering to long-term interruption constraints. As the NP-hard problem, we propose a Lyapunov-based Soft Actor-Critic (LSAC) algorithm to solve the optimization problem. Experimental results demonstrate that the suggested offloading architecture and algorithm have considerably enhanced performances in reducing interruption rate and delay.