Jointly Elements Allocation and Trajectory Design for Optical RIS-Assisted UAV Communications in Low-Altitude Obstructed Environments

Uncrewed aerial vehicles (UAVs) exhibit significant potential as mobile relays for establishing line-of-sight (LoS) links in free space optical (FSO) communications. However, providing reliable FSO communication to multiple users in obstructed environments remains a critical open problem. The integration of UAVs with optical reconfigurable intelligent surfaces (ORIS) further broadens the application prospects for FSO in multi-user environments with non-line-of-sight (NLoS) conditions. Therefore, this paper proposes an ORIS-assisted UAV relay system for FSO communications operating in obstructed low-altitude scenarios, aiming to provide services for multiple ground users. To maximize the average user throughput, a hungarian algorithm assisted proximal policy optimization (HA-PPO) method is proposed, which jointly optimizes ORIS elements allocation and UAV trajectory. Numerical simulation results demonstrate that the proposed approach effectively plans UAV trajectories to avoid obstacles and maintain LoS links, while achieving higher average throughput than its counterparts.