The state and bias estimation problem of spacecraft attitude control system considering non-Gaussian noise characteristics is studied. Firstly, a spacecraft attitude control system model with bias is established, and the expression for the non-Gaussian characteristic noise is given. Then, through l₁-norm Kalman Filtering (l₁-KF) algorithm and mathematical simulation, it is verified that the introduction of l₁-norm into the Kalman filtering algorithm can avoid the influence of non-Gaussian process noise. However, the l₁-KF algorithm cannot achieve simultaneous estimation of state and bias, and is only suitable for estimation of attitude stability control process. Therefore, on the basis of l₁-KF algorithm, considering the complex situation of time-varying moment of inertia and time-varying bias of the system, as well as real-time identification of time-varying moment of inertia, an l₁-norm Two-Stage eXogenous Kalman Filtering (l₁-TSXKF) algorithm is designed to estimate the state and bias of the spacecraft during attitude maneuver. The mathematical simulation results show that the l₁-TSXKF algorithm can reduce the influence of non-Gaussian characteristics of the system and obtain high precision attitude and bias estimation results quickly in the process of spacecraft attitude maneuver, which is conducive to the application of spacecraft in orbit.

An attitude maneuver control method based on a fully actuated system model is proposed for large-angle maneuvers of spacecraft in the presence of attitude forbidden region and disturbance torque. The attitude kinematics is developed on the 3-dimensional Special Orthogonal Group (SO (3) ), and the second-order fully actuated model of error attitude is established based on exponential coordinates, the corresponding Lie algebra of SO (3). The Barrier Lyapunov Functions (BLFs) based on SO (3) are introduced to ensure that the attitude constraints are obeyed when the values of BLFs are bounded. Finally, the Total Lyapunov Function (TLF) containing the BLFs is developed and analyzed. The nonlinear terms in the derivative of TLF are eliminated by using the sliding mode technique and the properties of parameter matrices in the second-order fully actuated model to get a concise control law. Numerical simulation results indicate that using the proposed control method, the spacecraft converges to the desired attitude without entering the attitude forbidden region, and the steady-state error and control energy consumption perform better than the existing methods.