UAV-Aided FSO Communication Alignment with YOLOv5 and PID Control Optimized by Deep Reinforcement Learning

Yulin Ding
College of Information Science and Engineering, Northeastern University, Shenyang 110819, Liaoning, China
Mingyi Gu
College of Information Science and Engineering, Northeastern University, Shenyang 110819, Liaoning, China
Yixiang Li
College of Information Science and Engineering, Northeastern University, Shenyang 110819, Liaoning, China
Tong Wei
College of Information Science and Engineering, Northeastern University, Shenyang 110819, Liaoning, China
Feiyue Wang
College of Information Science and Engineering, Northeastern University, Shenyang 110819, Liaoning, China
Songshan Tian
College of Information Science and Engineering, Northeastern University, Shenyang 110819, Liaoning, China

Abstract:

This paper presents a solution for aligning medium and short-distance FSO communication links using YOLOv5 for UAV detection and DDPG for PID parameter tuning. Implemented and verified through simulations and physical tests, the system achieves robust and stable alignment under various UAV trajectories. Results show enhanced communication quality.