Satellite orbit prediction based on two-stage filter

In order to improve the accuracy of the satellite orbit prediction, a Doppler/Celestial integrated orbit prediction method based on two-stage filter is proposed. Celestial prediction is a mature and stable navigation for satellite. But the position determination performance is not satisfied due to the low accuracy of horizon sensor. The Doppler prediction utilizes the Doppler radial velocity relative to the Sun as measurement information. However, this method can not obtain all state estimations. For two methods are complementary to each other, the two-stage filter is used here for fusing the prediction data from the Doppler prediction and CNS. Compared to the traditional celestial prediction method, the integrated prediction method can provide better prediction performance. The simulation results demonstrate the feasibility and effectiveness of this integrated prediction.