Hello dear scientists,

Nowadays, reinforcement learning is becoming a conventional approach for UAV guidance, navigation, and control systems, with various applications in path planning, path following, collision avoidance, etc.

I have studied several research papers in this field, and I can declare that none of them have employed RL for stability control in this control plant (UAV). In other words, authors have assigned this responsibility to other control approaches (why?), such as PID, SMC,...

The question is, Can Reinforcement Learning be reliable for UAV stability Control?