SINS/CNS Integrated Navigation With NFOV Star Tracker Based on Virtual Image Construction

A narrow-field-of-view (NFOV) star tracker is well suited for all-day detection scenarios, making it highly applicable to the strapdown inertial navigation system/celestial navigation system (SINS/CNS) integrated navigation system. However, analysis of the output characteristics of NFOV star trackers reveals that an NFOV star tracker cannot calculate the roll angle. Furthermore, image errors of the NFOV star tracker are identified as a significant factor that directly impacts the performance of attitude correction in integrated navigation systems. To address these issues, an SINS/CNS integrated navigation with an NFOV star tracker based on virtual image construction is presented. Specifically, a gyro-aided high-precision virtual image construction is proposed, based on which the NFOV star tracker's optimal tri-axis attitudes are estimated. In addition, an image error model for the NFOV star tracker is established. Moreover, a calibration method of tri-axis installation angles between the inertial measurement unit and the NFOV star tracker based on multistar observations is developed. Simulations and field tests demonstrate the ability to compensate for attitude misalignment angles and improve image detection accuracy compared with the other two methods, affirming the feasibility and efficacy of the proposed approach.