基于扩展卡尔曼滤波的星敏感器在轨几何标定

On-orbit Geometric Calibration of Star Tracker Based on EKF

星敏感器在轨几何标定是消除其系统测量误差,提高姿态确定精度的有效手段之一。目前利用星对角距相等进行星敏感器在轨标定时,大多采用最小二乘求得全局最优的静态待标定参数,但没有完全反映出恒星相机工作期间受外太空环境变化导致内部参数发生的微小动态变化。为了进一步提高星敏感器参数的在轨标定精度,文章在对星图识别结果预处理的基础上,引入扩展卡尔曼滤波对当前星图时刻对应的相机参数进行实时标定,动态获取当前最优参数值。利用该文所提方法对“资源三号”01/02星国产星敏感器进行在轨标定,试验结果表明所提方法能够有效标定当前星敏感器的参数误差,其精度优于传统的最小二乘标定方法,验证了该方法具有更好的适应性和可靠性,为星敏感器的在轨几何标定提供了一种技术参考,可用于卫星在轨实时标定或地面卫星姿态后处理。

The geometric calibration of star tracker on orbit is one of the effective means to eliminate the systematic measurement error and improve the accuracy of attitude determination. As to the present calibration method with star-pair angular distances, the least square method is mostly used to obtain the globally optimal static parameters to be calibrated, which does not fully characterize the small dynamic changes of internal parameters caused by the variations of outer space environment during the operation of star cameras. In order to further improve the on-orbit calibration accuracy of star tracker parameters, based on the preprocessing of the star map recognition results, this paper introduces Extended Kalman Filter (EKF) to calibrate the camera parameters corresponding to the current star map moments in real time and dynamically obtain the current optimal parameter values. Using the method proposed in the article, the ZY-3 01/02 domestic star tracker is calibrated on orbit. The test results show that the proposed method can effectively calibrate the parameter error of the current star tracker, with better accuracy, adaptability and reliability than the traditional least square method. The proposed method can be used for real-time satellite orbit calibration or ground satellite attitude post-processing, providing a technical reference for on-orbit geometry calibration of star tracker.