Task Offloading and Resource Allocation for MEC-Assisted UAV Using Meta-Reinforcement Learning

The present study examines the task offloading and resource allocation problem for a MEC-assisted UAV system. The objective is to minimize energy consumption and system latency, whilst also considering load balancing among the UAVs. In this paper, we propose a model-agnostic meta-reinforcement learning algorithm (MAML-DDPG) to address the optimization problem, enabling UAVs’ decision-making to quickly adapt to the time-varying optimization objectives. Simulation results show that, compared to existing methods, the proposed approach can rapidly adapt to dynamic optimization objectives while effectively reducing both the energy consumption of the UAVs and system latency while satisfying the load balancing requirements.