Research on Star Tracker On-orbit Low Spatial Frequency Error Compensation

Home » channel02 » Research on Star Tracker On-orbit Low Spatial Frequency Error Compensation
Research on Star Tracker On-orbit Low Spatial Frequency Error Compensation

Research on Star Tracker On-orbit Low Spatial Frequency Error Compensation

An on-orbit low spatial frequency error compensation method has been proposed to restrain low spatial frequency error brought by the difference between on ground calibration and on-orbit environment.Firstly,principle points,initial value of focal length and distortion coefficients are calculated during onground calibration.Then,focal length is updated by extended Kalman filter based on angle distance error minimization criterion using selected star pairs during on-orbit calibration.Faster convergence rate and better robustness has been validated by simulation.Sky tests data and on-board data proved that the mean value of angle distance error can be decreased by more than 90% and the low spatial frequency error decreased by more than 40%.

 

This article summarizes the in orbit calibration methods for star sensors and believes that the star diagonal distance provided by the star catalog is the most ideal benchmark for in orbit calibration. Unlike previous methods that optimize parameters such as focal length, principal point, and distortion simultaneously, this article believes that focal length is the main factor in in orbit changes. A low-frequency error in orbit compensation method combining ground calibration and astronomical calibration is proposed, which uses the filtered star diagonal distance to only perform one-dimensional filtering on focal length, The main point and distortion still use ground calibration parameters to enhance the reliability of in orbit calibration and simplify the problem of in orbit calibration. Simulation analysis shows that compared to the multi parameter simultaneous optimization method, this method has a faster convergence speed and is conducive to maintaining the stability of the filter. Multiple ground observation experiments have shown that this method can reduce the average star diagonal distance error to nearly zero, significantly reducing the low-frequency error of the star sensor’s field of view period, The reduction exceeds 40%. After analyzing the ground observation and in orbit flight data of high-precision star sensors of the same model, the focal length after in orbit calibration not only deviates from the initial value, but also adapts to the orbital environment, effectively compensating for low-frequency errors caused by factors such as launch vibration, changes in space thermal environment, and differences in star targets between heaven and earth, thereby improving the accuracy of star sensors.

Send us a message,we will answer your email shortly!

    Name*

    Email*

    Phone Number

    Message*