针对卫星姿控系统出现执行机构故障,提出一种基于迭代学习-未知输入观测器(iterative learning-unknown input observer,IL-UIO)的鲁棒故障重构方法。首先,考虑卫星出现空间干扰力矩、模型不确定性以及陀螺漂移,建立小角度机动时的非线性姿控系统模型。其次,采用UIO干扰解耦原理和H!控制思想,设计IL-UIO估计卫星姿态欧拉角和角速度的同时,利用IL算法实现执行机构鲁棒故障重构。并利用Lyapunov稳定性理论证明了IL-UIO稳定性和动态故障偏差最终有界,通过线性矩阵不等式(linear matrix inequaliry,LMI)工具箱求解了观测器部分参数矩阵。最后,建立卫星闭环姿控系统并进行仿真,仿真结果验证了此方法的有效性。 Aiming at the failure of actuator in satellite attitude control system, a robust fault reconstruction method based on iterative learning-unknown input observer (IL-UIO) is proposed. Firstly, the model of nonlinear attitude control system with small-angle maneuver is established considering space disturbance torque, model uncertainty and gyro drift of satellite. Secondly, IL-UIO was used to estimate the Euler angles and angular velocities of the satellite attitude by UIO interference decoupling principle and H! Control theory, and IL algorithm was used to realize the robust fault reconfiguration of the actuator. Finally, the Lyapunov stability theory is used to prove the ultimate boundedness of IL-UIO stability and dynamic fault tolerance. The partial parametric matrix of the observer is solved by a linear matrix inequaliry (LMI) toolbox. Finally, the satellite closed-loop attitude control system is established and simulated. The simulation results verify the effectiveness of this method.