The invention discloses a
star sensor calibration method based on a pitching rolling and rotating table and nonlinear optimization. According to the method, firstly, a
star sensor calibration
system based on the pitching rolling and rotating table and nonlinear optimization is built, wherein the
system comprises a parameter initial value
estimation module, a parameter recording module, a parametersolving module, a parameter optimization module and a parameter determining module. The method sequentially comprises the following steps of estimating the
focal length of the
star sensor and the initial value of the main point; recording the
mass centric coordinates of the star
point light spots in a star sensor measuring coordinate
system and the
direction vector of parallel light in the star sensor measuring coordinate system after the rotation; solving the angle distance cosine value of the star
point light posts in different coordinate systems; building optical parameters of a target function optimized star sensor; judging whether the
algorithm meets the iteration termination conditions or not through the star
point light spot re-projection errors for determining the final optimization parameters. The method has the advantages that the optical parameters of the star sensor are calibrated; meanwhile, the
rotation matrix between the pitching rolling and rotating table and the starsensor can be eliminated.