Research On X-ray Pulsar-based Navigation Method For Deep Space Exploration

Deep space exploration is another important component of China’s space strategy after satellite applications and manned spaceflight.Autonomous navigation technology for spacecraft is of great significance for improving the autonomous viability of deep space probes and alleviating the burden of ground measurement and control systems.As a new type of autonomous navigation technology for spacecraft,X-ray pulsar navigation is an important support for deep space exploration,spacecraft autonomous operation management,and near-Earth space complete navigation.As a hot spot in international deep space exploration,Sun-Earth L2 point has important practical application and scientific exploration value.Based on the application of X-ray pulsar navigation technology in the orbit of Sun-Earth L2 point,this paper conducts research on rapid simulation methods for observing X-ray pulsar signals at spacecraft,absolute and relative navigation methods for X-ray pulsars,etc.The main contents of the paper include:1.Aiming at the problem that existing simulation methods for observing pulsar signals at spacecraft are inefficient and unsuitable for simulating observation data at deep space orbit with long observation periods,a method for simulating observation signals based on the relationship between photon arrival time and phase at spacecraft is proposed.Firstly,the photon arrival phase sequence at the spacecraft is simulated,and then the expressions of the frequency and derivative of the pulsar signal observed at the spacecraft are derived.The relationship between the photon arrival time observed at the spacecraft and the corresponding phase is constructed.The transformation from photon phase to time is realized,avoiding complex processes such as iterative computation and real-time calculation of the photon arrival rate function at the spacecraft.Finally,a numerical simulation example is designed.The proposed simulation method is validated from multiple aspects such as pulse profile recovery,rotation frequency and derivative testing,comparison of measured data,and comparison of existing simulation methods.The results of simulation speed test indicate that the proposed simulation method is insensitive to changes in pulsar flow parameters and frequency parameters,and the simulation speed can be increased by up to three orders of magnitude compared with existing simulation methods.2.Aiming at the problem that preset pulsar observation periods cannot ensure the minimum navigation observation estimation error at the filtering time in navigation,an X-ray pulsar absolute navigation method based on optimal observation period estimation is proposed.In this paper,the observation period corresponding to the minimum navigation observation estimation error is defined as the optimal observation period.The relationship between the estimated mean square error of navigation observations and the observation period at the filtering time when observing three pulsars simultaneously is derived.Based on the minimum mean square error estimation criterion,the optimal observation period estimation formula is given,and a navigation algorithm based on the optimal observation period estimation is designed.The optimal observation period estimation formula and navigation algorithm are verified by simulation.The results show that the relative error of the optimal observation period is around 6%,and the navigation accuracy of the navigation method based on the optimal observation period estimation is improved by 45.9% compared to the navigation method with preset observation periods.3.Aiming at the problem of large relative position estimation error at the initial filtering time in relative navigation,a relative navigation method combining low order and high order relative position error propagation models is proposed.Based on the relative position error propagation models of different orders,the relative navigation observation equations and observation estimation models of X-ray pulsars are derived,and the relative navigation results using different order models are compared.The simulation results show that using higher order relative position error propagation models can achieve higher navigation accuracy at the initial stage of filtering.Combining the relative navigation results obtained using a higher order model at the initial stage of filtering with the relative navigation results obtained using a lower order model at other filtering stages,a relative navigation method combining a lower order and a higher order relative position error propagation model is designed.Simulation results show that through the combined method,the relative position estimation error at the initial stage of filtering is significantly reduced,and the fluctuation of the convergence curve becomes smaller;the relative navigation based on the combination of the first order and second order relative position error propagation models is more suitable for determining the relative state of the formation spacecraft at Sun-Earth L2 point.4.In order to solve the problem that the preset observation period of X-ray pulsars cannot guarantee the minimum estimation error of relative navigation observations in the prior assisted X-ray pulsar relative navigation,a prior assisted X-ray pulsar relative navigation method based on the optimal observation period estimation is proposed.The theoretical lower limit of the estimated variance of the phase deviation is derived when prior information is combined with the X-ray pulsar signal.Based on the lower bound of this theory,the relationship between the mean square error of the relative position estimation at the filtering time and the observation period is established when three pulsars are observed at the same time.A numerical solution for the observation period that minimizes the estimated mean square error of relative position estimation at the filtering time is designed,which can achieve the estimation of the optimal observation period for relative navigation.The simulation verifies the effectiveness of the proposed numerical solution for the optimal observation period and the relative navigation algorithm.The results show that the relative error of the optimal observation period for relative navigation with prior assistance is about10% compared to the relative navigation method with preset observation periods,the relative position estimation error of formation spacecraft is significantly reduced by estimating the optimal observation period under the same observation information.