基于自适应滤波的光谱畸变误差抑制方法

Suppression Method for Measurement Error Caused by Spectral Distortion Based on Adaptive Filtering

为降低天文光谱畸变误差对多普勒测速导航精度的影响,设计结合非线性Sage-Husa噪声估计器及抗差扩展卡尔曼滤波器(Robust Extend Kalman Filter,REKF)的自适应滤波算法。当系统模型可靠时,抗差滤波能够通过预测残差判断异常量测并降低其权重;当系统模型噪声先验信息不准确时,通过Sage-Husa噪声估计器估计系统噪声协方差阵Q阵,以保证抗差滤波的效果。此外,结合多普勒测速导航及X射线脉冲星导航进行组合导航,以提高位置估计精度。仿真结果表明,该算法能够在系统模型噪声先验信息不准确的情况下有效控制光谱畸变造成的量测误差对导航精度的影响。

In order to reduce the effects of the astronomical spectral distortion error on the accuracy of Doppler velocimetry navigation,an adaptive filtering algorithm is designed along with the nonlinear Sage-Husa noise estimator and the robust extended Kalman filter(REKF).When the system model is reliable,the robust filter can justify the abnormal measurement by predicting the residuals and reduce its weight.When the prior information of the system model noise is inaccurate,the system noise covariance matrix Q will be estimated by the Sage-Husa noise estimator to ensure the accuracy of the robust filter.In addition,the X-ray pulsar navigation is combined with the Doppler velocimetry navigation to improve the accuracy of position estimation.The simulation results show that the algorithm can effectively control the effects of the measurement error caused by spectral distortion on the navigation accuracy when the prior information of the system model noise is inaccurate.