Multirate iterative learning disturbance observer with robustness to frequency deviation for high precision attitude stabilization of flexible spacecraft subject to complex disturbances

• Published in: Aerospace science and technology Vol. 96; p. 105583
• Main Authors: He, Tongfu, Wu, Zhong
• Format: Journal Article
• Language: English
• Published: Elsevier Masson SAS 01.01.2020
• Subjects: Attitude control; Disturbance observer; Flexible spacecraft; Iterative leaning; Periodic disturbances; Disturbance observer; Iterative leaning; Attitude control; Periodic disturbances; Flexible spacecraft
• ISSN: 1270-9638; 1626-3219
• DOI: 10.1016/j.ast.2019.105583

High precision attitude stabilization of spacecraft is essential for the operation of precision instruments onboard. However, attitude vibrations of spacecraft are inevitably induced by complex disturbances. Disturbance observer based control scheme (DOBC) is a promising anti-disturbance technique, while it is restricted by the inadequate estimation performance of conventional disturbance observers for complex disturbances when taking structural complexity and noise sensitivity into account. To achieve high performance attitude stabilization of spacecraft subject to complex disturbances, a multirate iterative learning disturbance observer (MILDO) based controller is designed in this paper. Firstly, an augmented disturbance model is constructed in view of the periodicities of disturbances with known fundamental frequencies. Then, a generic proportional-derivative type MILDO (PD-MILDO) is designed by using previous disturbance estimates to update current estimates. The gain tuning method and the disturbance estimation performance of PD-MILDO are also presented. It is demonstrated that accurate disturbance estimation for low-frequency disturbance and multiple periodic disturbances can be achieved by PD-MILDO. Besides, the robustness of PD-MILDO to undesired frequency deviation can be significantly improved by introducing the proportional-derivative correction terms. Based on MILDO, a feedback controller is then designed to stabilize the spacecraft attitude. In the concept of DOBC, the controller and observer can be tuned independently. Finally, the effectiveness of the proposed control scheme is verified by simulations.