KINEMATIC ORBIT DETERMINATION OF GEO SATELLITE BASED ON KALMAN FILTERING
He Kaifei 1) ; and Xu Tianhe 2,3)
1)School of Geology Engineering and Geomatics, Chang’an University, Xi’an 7100542)School of Geodesy and Geomatics, Wuhan University, Wuhan 4300793)Xi’an Research Institute of Surveying and Mapping, Xi’an 710054
Abstract Considering the features of the motion of GEO( Geostationary Earth Orbit ) satellite, Kalman filtering by using constant velocity model as state equation is proposed. When there exists systematic errors, forward and backward filtering is used, and when not, forward filtering. And the idea behind forward and backward filtering is that, firstly, using forward filtering estimates systematic error parameters, then the values obtained are used to do backward filtering. The simulation results show when there exists only random errors in observations, the use of orbit determination by forward Kalman filtering has a relatively high accuracy, but once system errors are involved, the accuracy will decrease sharply. The forward Kalman filtering considering systematic error parameters can weaken the impacts of systematic errors to a certain degree; while the application of forward and backward Kalman filtering will further weaken the influences of systematic errors and improve the accuracy of orbit determination. The results also show that orbit determination by Kalman filtering can comparatively exactly estimate the running velocity of GEO satellite.
