X-ray Pulsar-based Spacecraft Autonomous Navigation

The autonomous positioning system for spacecraft can determine the spacecraft's positions without ground stations. The autonomous navigation not only reduces costs, but also is fit for deep explore. In recent years, the autonomous positioning for spacecraft based on x-ray pulsar has been a hot research subject for spacecraft technology. In this paper, we research the key technologies of pulsar navigation. The main contents are as follows.At pulsar signal preprocessing, a pulsar signal de-noising algorithm and a pulsar signal recognition algorithm are proposed. A pulsar signal de-noising algorithm based on empirical mode decomposition is proposed. The empirical mode decomposition method decomposes pulsar signal into a group of intrinsic mode functions, and the adaptive threshold is used to deal with intrinsic mode functions. A pulsar signal recognition algorithm based on bispectra in wavelet domain is proposed. Firstly, the bispectra of wavelet coefficients is calculated after signal is decomposed by wavelet transform. Then, the bispectral features of low and high frequency coefficients are extracted by the principal component analysis and the selected bispectra respectively. Finally, the minimum distance classifier is used to classify pulsar signals. The simulation results have demonstrated the feasibility and effectiveness of two pulsar signal.Three single pulsar/celestial integrated navigation methods are proposed. Considering that technology ability and money is finite, "Single Pulsar for Autonomous Navigation" can be implemented. Single pulsar navigation system is completely unobservable. And the celestial navigation system is completely observable, but its accuracy of position determination is low. For two methods are complementary to each other, three single pulsar/celestial integrated navigation methods are proposed. The navigation method based on federated Unscented Kalman Filter (UKF) can fuse the navigation information from single pulsar navigation and celestial navigation subsystems. An integrated navigation method based on multiple model adaptive estimation is proposed. An UKF based on multiple model adaptive estimation is designed to solve the problem of nonlinear navigation system and the uncertainty of the pulse time of arrival noise. The integrated navigation based on UKF/H∞filter is suitable to deal with clock drift. The epoch-difference method is adopted to eliminate the clock drift, but it makes the pulsar navigation measurement noise become coloured. Two-level filter, which is composed of UKF and H∞filter in series connected, is designed The celestial navigation and pulsar navigation subsystems utilize UKF and H∞filter respectively. Simulation results demonstrated three navigation methods can deal with the faced problem, and obtain high positioning accuracy.In order to reduce the impact of the pulsar direction error on the navigation system performance, a novel X-ray pulsar navigation technique is proposed. Through analyzing the system bias caused by the pulsar direction error, it can be seen that the system bias is slowly time-varying. Based on the analysis result, the augmented state unscented Kalman filter, in which the system bias is treated as the augmented state, is designed here to deal with the system bias and estimate spacecraft's positions and velocities. The augmented state unscented Kalman filter-based navigation method for spacecraft is more accurate than the method based on unscented Kalman filter in the presence of the pulsar direction error.A Doppler/pulsar integrated navigation method is proposed. The Doppler navigation utilizes the Doppler radial velocity relative to Sun as measurement information, and this method is completely unobservable. The pulsar navigation system using more than 3 pulsars is completely observable, but the filtering period is long. For two methods are complementary to each other, a federated UKF is utilized to fuse the information from the pulsar navigation and the Doppler navigation. Compared to the pulsar navigation, the integrated navigation method can provide better navigation performance.