Multi-Agent Deep Reinforcement Learning for Trajectory Design and Power Allocation in Multi-UAV Networks | IEEE Journals & Magazine | IEEE Xplore