Delay-Aware and Energy-Efficient Task Offloading for IoV: A Thread-Cooperative A3C Approach

The development of the Internet of Vehicles (IoV) has led to vehicles facing frequent computation-intensive tasks. Multi-access edge computing (MEC) has become an effective solution to reduce the computing load and enhance service performance for IoV. In this letter, we aim to optimize the serving delay and energy consumption of intelligent vehicles (IVs) in MEC-assisted IoV. We formulate an optimization problem to reduce the computing load of IVs taking into account the vehicular mobility, wireless channel conditions, and task delay requirements. Considering the dynamics of the optimization problem, we propose an multi-threaded interactive A3C (MIA3C) algorithm to make fast and accurate online decisions. To our knowledge, this is the first attempt to address the task offloading and resource allocation problem in the dynamic vehicle mobility and wireless transmission conditions. The simulation results show that the proposed MIA3C holds a good convergence rate and the proposed scheme can achieve better performance in terms of energy consumption and task offloading rate compared with existing schemes.