Quick-response attitude takeover control using multiple servicing spacecraft based on inertia properties identification

Abstract

Listen

Translate article

• Abrupt change of inertia is a critical obstacle to attitude takeover control. • Identification of inertial parameters is updated with only single sampling data. • Integrated method can quickly and accurately respond to variable inertia properties. • Consumptions of computing and storage are greatly reduced thus being more practical. • Stable and high-precision attitude control can be actualized even at attached moment. Multi-spacecraft is essential to on-orbit servicing (OOS) which leads to spacecraft life prolongation thus reducing space debris. However, the abrupt change of inertia parameters caused by the attachment of each servicing spacecraft is a critical obstacle to the stable and high-accuracy attitude takeover control of a target spacecraft during the entire servicing process using multiple servicing spacecraft. Therefore, a quick-response attitude takeover control method based on inertia properties identification is proposed in this paper. Combining the Euler dynamical equations of combined spacecraft with space environment moment model, a novel iterative identification equation integrating inertia matrix is established to eliminate ill-conditioned identification and realize the precise identification of inertia parameters using only single sampling data for each update. With the quickly and precisely estimated inertia parameters, a Lyapunov-based attitude controller and an optimal torque allocation method are designed to actualize the high-accuracy and globally stable attitude takeover control with the minimum energy consumption. The proposed method can actualize the quick-response and stable attitude takeover control with low computational cost, even at the moment of attachment of servicing spacecraft. Hence, it is greatly appropriate for the OOS missions with multiple servicing spacecraft. The ground experiments and numerical simulations were conducted for demonstrating the feasibility and effectiveness of the proposed method. The experiment result indicated that the trajectory tracking error could converge to ± 0.02 ° and ± 0.02 ° / s and the parameters identification error was less than 5.5%.