Multi-Agent Deep Reinforcement Learning-Based Trajectory Planning for Multi-UAV Assisted Mobile Edge Computing

An unmanned aerial vehicle (UAV)-aided mobile edge computing (MEC) framework is proposed, where several UAVs having different trajectories fly over the target area and support user equipments (UEs) on ground. We aim to jointly optimize geographical fairness among all UEs, of each UAV' UE-load overall energy consumption UEs. The above optimization problem includes both integer continues variables it challenging solve. To address problem, a multi-agent deep reinforcement learning based trajectory control algorithm proposed for managing UAV independently, popular Multi-Agent Deep Deterministic Policy Gradient (MADDPG) method applied. Given UAVs' trajectories, low-complexity approach introduced optimizing offloading decisions show that our solution has considerable performance other traditional algorithms, in terms serving at