Performance evaluation of relative navigation algorithm during geostationary orbit long-range rendezvous

Long-range autonomous rendezvous is crucial to the precise placemen of geostationary collocation satellites. Based on the analysis of error signal characteristics of relative motion model during large-span remote rendezvous process, the error transfer mechanism is theoretically derived. Then the quantitative analysis model of mean and noise characteristic is proposed for the relative navigation based on an extended Kalman filter (EKF) algorithm. The analysis results show that there exists an optimal process noise level and corresponding optimal filter performance with respect to the determining distance range. The estimation error is positively correlated with the intersection distance, which can be used as the quantitative basis to design a global optimum filter. Besides, it also theoretically provides an approach to quantitatively evaluate the navigation performance of the EKF optimal filter during long-range rendezvous of geostationary orbit. Simulation results demonstrate that the analysis results are correct, and the relative velocity estimation error is 0.005 m/s (1σ), which satisfies the mission requirements of orbit long-range rendezvous.