Safe Reinforcement Learning for Transition Control of Ducted-Fan UAVs

Ducted-fan tail-sitter unmanned aerial vehicles (UAVs) provide versatility and unique benefits, attracting significant attention in various applications. This study focuses on developing a safe reinforcement learning method for back-transition control between level flight mode hover ducted-fan UAVs. Our enables transition with minimal altitude change time while adhering to the velocity constraint. We employ Trust Region Policy Optimization, Proximal Optimization Lagrangian, Constrained (CPO) algorithms controller training, showcasing superiority of CPO algorithm necessity The trajectory achieved using closely resembles optimal obtained via well-known GPOPS-II software SNOPT solver. Meanwhile, also exhibits strong robustness under unknown perturbations UAV model parameters wind disturbance.