Improved autonomous orbit determination algorithm using star sensor and earth sensor

Abstract

To solve the application difficulty and the low accuracy problem in the engineering field, an improved autonomous orbit determination method using star sensor and earth sensor was proposed. First, the observation variable was changed. That is, to adapt the star sensor standard outputs, the geocentric unit vector of the inertial coordinate system instead of the starlight angular distance was taken as the observation variable in the Kalman filter equations. Second, an error elimination section was added into the method. Constant errors were calculated to correct the measured geocentric vector, and the spike noises were eliminated by the fault-tolerant improvement method. Third, the improved autonomous orbit determination method with the new observation variable and an error elimination process was implemented based on the unscented Kalman filter method. The verification teats through in-orbit data of a medium-earth-orbit satellite show that the orbit determination error is in the level of kilometers, and the proposed method can be effectively applied in the engineering field. © 2018, Editorial Department of Journal of Chinese Inertial Technology. All right reserved.