Separation of Intrinsic and Extrinsic Parameters for Star Sensors Based on Feature Trajectories Imaging With Calibration System

The method of integrated modeling with intrinsic and extrinsic parameters for star sensor calibration based on star simulators and rotary tables faces the coupling problem between intrinsic and extrinsic parameters. A novel method of separating the intrinsic and extrinsic parameters for star sensors was proposed in this paper to resolve the problem. This paper analyzed and revealed the properties of the imaging feature trajectories of star sensors in the calibration system. Based on the properties, the initial alignment deviation between the star simulator and the rotary table was eliminated, and the principal point of the star sensor was separated. Furthermore, the accuracy of separating the principal point was improved by the optimized design and processing of elliptical trajectories. Finally, the remaining parameters were comprehensively solved. The simulations and physical experiments validated the effectiveness of the proposed method. Compared to the integrated modeling method, the proposed method separated the principal point, which exhibits the strongest coupling with extrinsic parameters. Its consistency was approximately improved from a level of 25 pixels to 5 pixels. Additionally, the consistency of other intrinsic parameters was improved by more than 30%.