Trajectory tracking and anti-disturbance control for aerial recovery drogues without flow angle measurements

• Published in: Applied mathematical modelling Vol. 145; p. 116105
• Main Authors: Liu, Yiheng, Wang, Honglun, Wang, Yanxiang, Zhu, Junfan, Fan, Jiaxuan
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.09.2025
• Subjects: Aerial recovery; Appointed-time prescribed performance control; Deep learning; Nonlinear dynamic; Trajectory tracking; Deep learning; Trajectory tracking; Aerial recovery; Appointed-time prescribed performance control; Nonlinear dynamic
• ISSN: 0307-904X
• DOI: 10.1016/j.apm.2025.116105

•A deep learning-based flow angle component prediction network (FPN) is proposed.•A “bridge” connecting the trajectory loop and attitude loop of the drogue is proposed.•The appointed-time prescribed performance control method is developed.•Auxiliary systems are developed to handle the input saturation problem.•The backstepping-based controllers are designed according to drogue 6-DOF dynamics. For the aerial docking problem, as one of the docking subjects, the active control of the drogue is very important. However, there are challenges such as multiple disturbances, inability to install accurate measurement sensors, and attitude constraints. Aiming at these problems, this paper proposes a trajectory anti-disturbance control method for the recovery drogue without flow angle measurements to effectively suppress the sensitivity of the drogue to multiple wind disturbances and provide favorable conditions for successful docking. First, a deep learning-based conversion method from flow angle to Euler angle under wind disturbance conditions is proposed. Then, an affine connection between the position loop and attitude loop is established, and the overall anti-disturbance control framework from attitude to position is formed to realize the high-precision control of drogue position under multiple disturbance. Furthermore, the appointed-time prescribed performance control method is developed to accurately constrain the Euler angles to meet the docking requirements. Finally, the stability of the proposed method is proven, and the effectiveness is demonstrated by abundant simulations.