The controller design of unmanned aerial vehicle (UAV) often faces challenges such as significant parameter uncertainty, external interference, existence of unmodeled dynamics and cross-coupling effect. An intelligent robust anti-disturbance control method based on the Deep Q Network (DQN) is designed, which effectively solves the difficult problem of UAV controller design. Firstly, the UAV longitudinal nonlinear model is decoupled. Secondly, a robust anti-disturbance controller is designed to solve the problem that the extended state observer in the automatic disturbance rejection control has an error in estimating the total disturbance without considering the change of control input parameters. Aiming at the strong requirements of UAV for tracking control convergence speed in complex environment, an intelligent and robust anti-interference controller is designed based on the DQN algorithm. Finally, simulation examples are used to verify that the designed controller can improve the tracking control accuracy and convergence speed of UAV, does not depend on the mathematical model of the aircraft, and has strong robustness.
Intelligent Robust Disturbance Rejection Control of UAV Based on Deep Q Network
Lect. Notes Electrical Eng.
International Conference on Autonomous Unmanned Systems ; 2023 ; Nanjing, China September 09, 2023 - September 11, 2023
Proceedings of 3rd 2023 International Conference on Autonomous Unmanned Systems (3rd ICAUS 2023) ; Kapitel : 51 ; 546-554
26.04.2024
9 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Intelligent Robust Disturbance Rejection Control via Deep Reinforcement Learning
British Library Conference Proceedings | 2022
|Intelligent Robust Disturbance Rejection Control via Deep Reinforcement Learning
Springer Verlag | 2022
|Robust error-based active disturbance rejection control of a quadrotor
Emerald Group Publishing | 2020
|System identification and model-based robust nonlinear disturbance rejection control
TIBKAT | 2021
|