考虑地球扁率误差的两步Kalman滤波星光导航方法

A starlight navigation algorithm based on two-step Kalman filter method with earth oblateness correction

为提升基于星敏感器和地球敏感器的卫星星光导航精度,提出了一种基于两步Kalman滤波的星光导航算法。通过一步星光导航滤波器预处理,得到了存在一定误差的卫星位置估计。将位置估计引入地球扁率误差修正算法,使地球敏感器地球扁率误差实现星上自主修正。而后,利用修正后的地球敏感器输出形成精确的星光角距观测信息,再次应用星光导航滤波器,得到了精度较高的卫星轨道信息。仿真结果表明,利用所提出的算法,地球敏感器敏感地心矢量精度为0.001°,卫星三轴位置精度为3000 m,卫星三轴速度精度为4 m/s。相比于传统无修正星光导航算法,精度提升了80%以上。

To improve the accuracy of the starlight navigation algorithm,a two-step Kalman filter algorithm is proposed.Through first-step star navigation filter preprocessing,a rough satellite position estimation is obtained.The position estimation is introduced into the earth oblateness error correction algorithm and the earth oblateness error in the earth sensor is corrected effectively and autonomously.Then,using the corrected earth sensor outputs to form accurate star angular distance observation information,the starlight navigation algorithm is reused and the navigation messages are precisely determined.The simulation results show that by the two-step Kalman filter algorithm,the accuracy of the geocentric vector measured by the earth sensor is 0.001°,the three-axis position estimation errors of satellite are 3000 m,and the three-axis velocity estimation errors of the satellite are 4 m/s.Compared with the traditional uncorrected starlight navigation algorithm,the accuracy has been improved by more than 80%.