Scholarlink

<p>This work is an investment in addressing the optimal trajectory tracking control problem for a quadrotor unmanned aerial vehicle (UAV) subjected to external disturbance. The proposed method is a combination of an Adaptive Dynamic Programming (ADP)-based controller and a nonlinear disturbance observer (NDO), ensuring the stability of the closed-loop quadrotor under the influence of unknown external disturbance and achieving near-optimal trajectory tracking control. Firstly, the tracking error dynamics are established by utilizing the feedforward terms to compensate for the known time-varying error information in the quadrotor dynamics. Then, a nonlinear disturbance observer (NDO) is designed to compensate for unknown external disturbances. By eliminating the influence of external disturbances and time-varying dynamics in the quadrotor system, the error model of the quadrotor is transferred to the nominal model. A critic-only learning structure is applied for both position and attitude controllers to learn the optimal feedback control policy. Finally, a flight simulation is given to illustrate the superior efficiency of the proposed control scheme. The simulation result shows that the proposed method lies in its ability to compute a smooth optimal control policy even under the influence of disturbances, while effectively mitigating the chattering effect.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Disturbance Observer-Based On-Policy Adaptive Dynamic Programming for a Quadrotor UAV with External Disturbance

  • 期刊论文

  • 发表日期：

    2025年8月19日

查看全文
  • Ngoc Trung Dang,
  • Quynh Nga Duong,
  • Gia Khiem Dinh

摘要

This work is an investment in addressing the optimal trajectory tracking control problem for a quadrotor unmanned aerial vehicle (UAV) subjected to external disturbance. The proposed method is a combination of an Adaptive Dynamic Programming (ADP)-based controller and a nonlinear disturbance observer (NDO), ensuring the stability of the closed-loop quadrotor under the influence of unknown external disturbance and achieving near-optimal trajectory tracking control. Firstly, the tracking error dynamics are established by utilizing the feedforward terms to compensate for the known time-varying error information in the quadrotor dynamics. Then, a nonlinear disturbance observer (NDO) is designed to compensate for unknown external disturbances. By eliminating the influence of external disturbances and time-varying dynamics in the quadrotor system, the error model of the quadrotor is transferred to the nominal model. A critic-only learning structure is applied for both position and attitude controllers to learn the optimal feedback control policy. Finally, a flight simulation is given to illustrate the superior efficiency of the proposed control scheme. The simulation result shows that the proposed method lies in its ability to compute a smooth optimal control policy even under the influence of disturbances, while effectively mitigating the chattering effect.