On-Orbit Joint Estimation Method of Attitude and Intrinsic Parameters for High-Precision Star Sensors

The intrinsic parameters of a star sensor are critical factors affecting its attitude determination accuracy. However, due to mechanical vibrations, temperature variations, and space radiation, these parameters may undergo long-term drift during on-orbit operation. High-precision star sensors impose stringent requirements on pointing accuracy. Therefore, periodic intrinsic recalibration is essential. In order to satisfy the high-precision requirements of specific tasks, this paper proposes an on-orbit joint estimation method of attitude and intrinsic parameters for star sensors. First, a joint estimation approach for attitude and focal length based on Singular Value Decomposition (SVD) is presented. This method provides reliable focal length and initial attitude for subsequent optimization. Second, a joint estimation framework based on theWahba model is developed for attitude and intrinsic parameters, the parameters iterative update formulas are derived. Third, the Cramér-Rao Lower Bound (CRLB) of the attitude-dependent calibration model is derived, providing a theoretical benchmark for the estimation accuracy. Finally, both simulation and physical experiments validate the effectiveness and reliability of the proposed method. Under the current small field-of-view (FOV) conditions for star sensors, the proposed method significantly outperforms conventional onorbit calibration methods. It achieves a twofold improvement in focal length accuracy and a 20% improvement in principal point accuracy. The parameter estimation bias has also been significantly reduced.