A Multi-source Data Fusion Navigation Method of Spacecraft with Limited GNSS Signals

Global Navigation Satellite system (GNSS) is initially designed to provide services to objects on the earth surfaces, or in the earth-orbits with altitude below 3000 km. However, previous researches show that GNSS receivers mounted on GEO satellites can receive GNSS signals and generate positioning and navigation results as well. Considering that in some harsh situations, such as bad observation geometry, weak signal acquisition, etc., the on-board GNSS receivers can only transfer pseudo-range data to GEO satellites, this paper proposed an autonomous navigation method in which the information from GNSS receiver, optical attitude sensors, and inter-satellite links are fused together: at first, the extended state equation and measurement equation of a GEO satellite with optical sensor’s errors are deduced; then, an extended Kalman filter is designed to evaluate and compensate optical sensor’s errors via constructed earth-centered vector bias; later, a sequential correction is carried out using inter-satellite link measurements; finally, the performance of proposed method is analyzed through Monte-Carlo simulation. It shows that the average distance and velocity evaluation errors of proposed method are around 0.2 km and 0.05 m/s individually; the performance can meet the requirements of survivability of a class of high-orbit, long-life and high-value satellites.