Digital Twin Assisted Task Offloading for Aerial Edge Computing and Networks

Considering the user mobility and unpredictable mobile edge computing (MEC) environments, this paper studies intelligent task offloading problem in unmanned aerial vehicle (UAV)-enabled MEC with assistance of digital twin (DT). We aim at minimizing energy consumption entire system by jointly optimizing terminal users (MTUs) association, UAV trajectory, transmission power distribution computation capacity allocation while respecting constraints mission maximum processing delays. Specifically, double deep Q-network (DDQN) algorithm stemming from reinforcement learning is first proposed to effectively solve MTUs association trajectory. Then, closed-form expression employed handle further addressed via an iterative algorithm. Numerical results show that our scheme able converge significantly reduce total compared benchmark schemes.