Installation angle calibration between the star sensor and gyro unit using the equivalent rotation vector transformation

Abstract

To improve the dynamic performance of the star sensor, different star image frames are correlated using the attitude-correlated frame (ACF) approach, and the attitude changes among these star image frames are measured by the strapdown gyro unit (GU). The GU consists of three orthogonally assembled gyros, while the accelerometer is excluded. Accurate calibration of the installation angles between the star sensor and the GU is essential for the ACF approach. An installation angle calibration method using the equivalent rotation vector transformation between the star sensor and the GU is proposed in this work. Through three to five maneuverings of the star sensor/GU system, the equivalent rotation vector transformation of the observed vector is performed by the installation matrix, and the optimal values of the installation angles are estimated. Simulations and experiments are designed and conducted, and both verify the proposed calibration method. The calibration results in the experiment show that the standard deviation of the installation angles in each measurement data group is less than 5 arcseconds and the difference between different data sets is less than 14 arcseconds, both for the three axes, which are accurate enough for the ACF approach and other applications.