Attitude Control of Fixed-wing UAV Based on DDQN

In this paper, the Double Deep Q-Learning (DDQN) which is one of the deep reinforcement learning (DRL) algorithms, is used to train an agent to control the pitch channel attitude of a fixed-wing unmanned aerial vehicle (UAV) in the laboratory. Non-linear attitude dynamics model of the UAV’s pitch channel and the corresponding Markov decision process (MDP) have been established. On this basis, agent training and testing are carried out. The results show that the trained agent has a certain attitude control ability, which means the research direction has a certain value and potential.