【摘要】 针对在轨服务机动过程中卫星推力器、飞轮执行机构性能衰减问题，提出了一种基于滚动时域估计的模型预测容错控制方法。首先给出了基于对偶四元数的卫星姿轨一体化误差动力学模型，基于此设计了适用于姿轨一体跟踪控制的显式模型预测控制策略。然后，考虑在轨卫星推力器及飞轮执行机构的退化失效故障，设计由历史系统状态与控制序列驱动的滚动时域估计器(MHE),对故障执行机构的输出效率进行快速估计并反馈给控制器，在线实时调整控制器参数，最终完成了执行机构故障下适用于姿轨协同机动过程的模型预测容错控制设计。仿真结果表明，在设计的模型预测容错控制的作用下，系统能够快速准确地估计执行机构的故障系数，实现对期望姿轨状态的稳定跟踪控制。更多还原

【Abstract】 Considering the performance degradation of actuators such as satellite thrusters and flywheels during on-orbit servicing maneuvers, a fault-tolerant model predictive control method based on moving horizon estimation is proposed. Firstly, an integrated dynamics error model of satellite attitude and orbit control based on dual quaternion is presented. Based on the model, an explicit model predictive control strategy is designed for pose integrated tracking control. Additionally, considering the degradation fault of thrusters and the flywheels of on-orbit satellite, a moving horizon estimator(MHE) is designed to quickly estimate the output efficiency of the fault actuator and feed it back to the controller to adjust the controller parameters online and in real time, in order to accomplish the tracking control of the target. The simulation results show that the system can online estimate the fault coefficients of the actuator quickly and accurately and realize the stable tracking control during pose maneuvering process.更多还原