Research On Coordinated Attitude Control Approach Of Satellite Formation Flying And Compensation Of Input Delay

The recent decades have witnessed active studies in attitude coordination control of spacecraft formation flying (SFF) for its significant roles in missions such as synthetic-aperture radar imaging and space-borne interferometry. However, for the sake of the existence of the detrimental factors such as model uncertainties, external disturbances and time delays in communication links, the performance of coordination control is not always satisfactory. To solve this problem, the paper focuses on the theory and methods for attitude coordination control of SFF. The main contents of this dissertation are as follows:The coordinate attitude control problem for SFF in the Leader-follower mode is investigated. Based on the traditional tracking control methods for a single spacecraft, an attitude coordination controller for SFF is proposed, and the stability of the controller is analyzed. Considering the model uncertainties and external disturbances, a robust controller is applied, according to the corollary of Barbalat's lemma, the convergence of the system is proven theoretically.The study on coordinate attitude control for SFF in the behavior-based mode is also performed. Considering the attitude of the near-by spacecraft, an attitude coordination controller for SFF is designed, and the stability of the control system is analyzed by Lyapunov theory. In addition, based on the controller above, another controller which takes the external disturbances into consideration is proposed for behavior-based SFF, and also the convergence of the closed loop system is proven by the corollary of Barbalat's lemma.There always exist time delays in communication links, as for this problem, Artificial Neural Network (ANN) is used for its non-linear fitting and prediction capabilities. Focusing on the situations that the input attitude information is invariant, an offline time delays compensator is designed based on linear ANN. While, considering about the situations that the input attitude information is variant, by limiting the sample, refreshing the sample concurrently and setting threshold value of errors, an online time delays compensator is presented.