Attitude Control of Hypersonic Vehicle based on Reinforcement Learning

In this paper, the attitude control problem of hypersonic vehicle is studied. The DDQN (double deep Q network) algorithm is used to realize the online adaptive adjustment of the control parameters of the PID (proportional-integral-derivative) controller. Firstly, the dynamic model of hypersonic vehicle is established. Then, the implementation process of parameter adjustment using DDQN algorithm is described. By building two neural networks, the main network is used to determine the action to be performed, and the target network is used to determine the action value, so as to realize the adaptive adjustment of the control parameters. Finally, the effectiveness of the proposed method is verified by simulation experiments.