Recent Studies on Deep Reinforcement Learning in RIS-UAV Communication Networks

Unmanned aerial vehicle (UAV) and reconfigurable intelligent surface (RIS) technologies have recently been identified as enablers for future wireless networks. Deep reinforcement learning (DRL) is also a potential technique for optimizing performance in dynamic and complex networking environments. In this paper, we examine the state-of-the-art studies on DRL utilization in RIS-UAV communication systems concerning their objectives, optimization parameters, deployment scenarios, and DRL methods. In addition, we emphasize research challenges and directions that can be addressed to improve RIS-UAV networks.