This paper proposes a novel centralized training and distributed execution (CTDE)-based multi-agent deep reinforcement learning (MADRL) method for multiple unmanned aerial vehicles (UAVs) control in autonomous mobile access applications. For the purpose, a single neural network is utilized in centralized training for cooperation among multiple agents while maximizing the total quality of service (QoS) in mobile access applications.
Acknowledgments. This research was funded by National Research Foundation of Korea (2022R1A2C2004869). Chanyoung Park and Haemin Lee equally contributed to this work (first authors). Soohyun Park is a corresponding author (soohyun828@korea.ac.kr).