利用预测滤波法估计小卫星姿态角速度

On Better Predictive Filtering for Gyroless Microsatellite Attitude Rate Estimation

预测滤波法具有可直接处理线性或非线性模型、实时预测模型误差的优点。结合小卫星姿态运动学方程和预测滤波法 ,建立了简单的线性估计模型 ,实时估计姿态角速度 ,并保证了估计的最优性。仿真结果表明 ,对姿态角速度的估计精度较高 ,鲁棒性好。

Past research [3,4,6] that used PF (predictive filtering) to obtain attitude rate estimation of gyroless microsatellite had to rely on experience to select model error weighted matrix. We propose an estimation algorithm for determining such a matrix. This algorithm is a very simple linear estimation algorithm and requires only the use of angle sensors. Eq.(16) can be used to determine the weight w of the model error weighted matrix; thus the model error weighted matrix is determined. This method's merits involve direct use of system models (linear or nonlinear) and real time prodiction of model error. This algorithm can guarantee the optimality of estimation. Simulation results, as shown in Figs.1, 2, and 3, show that our PF algorithm is suitable for any attitude angle rate estimation as both the accuracy and robustness are good.