Research on the filter algorithm for integrated autonomous navigation based on angle and velocity measurement in deep space

In order to reduce the influence of the constant measurement error, slow time-varying measurement error and spectral distortion measurement error caused by global and local activities of celestial bodies on the Doppler navigation, an augmented state adaptive cubature Kalman filter is proposed. The slow time-varying error can be regarded as a constant error in a short time and is estimated as one of the state variables and compensated in the measurement. The measurement noise covariance matrix is modified adaptively by Sage-Husa noise estimator to reflect the characteristics of noise and reduce the influence of spectral distortion measurement error. In addition, the celestial angle information is combined with the velocity information to improve the accuracy of autonomous navigation system. The simulation results show that the algorithm can estimate the constant and slow time-varying measurement error accurately and is robust in controlling the measurement error caused by spectral distortion.