Robust Computation Offloading and Trajectory Optimization for Multi-UAV-Assisted MEC: A Multi-Agent DRL Approach

For multiple Unmanned-Aerial-Vehicles (UAVs) assisted Mobile Edge Computing (MEC) networks, we study the problem of combined computation and communication for user equipments deployed with multi-type tasks. Specifically, consider that MEC network encompasses both uncertainties, where partial channel state information inaccurate estimation task complexity are only available. We introduce a robust design accounting these uncertainties minimize total weighted energy consumption by jointly optimizing UAV trajectory, partition, as well resource allocation in multi-UAV scenario. The formulated is challenging to solve coupled optimization variables high uncertainties. To overcome this issue, reformulate multi-agent Markov decision process propose proximal policy Beta distribution framework achieve flexible learning policy. Numerical results demonstrate effectiveness robustness proposed algorithm multi-UAV-assisted network, which outperforms representative benchmarks deep reinforcement heuristic algorithms.