Research On Autonomous Celestial Navigation Algorithm For Lunar Probe’s Earth-moon Transfer Segment

The complete success of the Chang’e 5 probe mission in 2020 has brought a perfect end to the first phase of my country’s lunar exploration project,and also brought more vision for future lunar exploration missions.Lunar exploration missions face difficulties such as complex environments,long flight times,and difficulty in ensuring full autonomy.Traditional ground-based measurement and control navigation methods have problems such as limited measurement and control arcs,low real-time performance,large signal attenuation,and extended measurement and control time.The celestial navigation methods represented by starlight navigation and X-ray pulsar navigation have the characteristics of high concealment,strong autonomy,strong anti-interference ability,and high accuracy.Therefore,this article will conduct research on two celestial navigation methods,and the design can be applied to the moon The probe’s fully autonomous navigation system.Aiming at the problem of state modeling for lunar probes flying in the Earth-Moon transfer orbit,from the circular restricted three-body model,the circular restricted four-body model and the complete model,the CR4 BP that takes into account the calculation efficiency and accuracy is selected.Posture model choose Euler angle posture motion model,and carry out simulation analysis of posture orbit motion.As the most mature celestial navigation algorithm at present,starlight navigation is to identify the position,speed and attitude information of the carrier by matching and identifying the observation star detected by the star sensor with the navigation star with a known position in the universe.In view of the celestial navigation of the lunar probe,carry out the research on the ground calibration method of the star sensor,simulate the data acquisition and image processing process of the star sensor under low and high dynamics,calculate the attitude information of the probe.The starlight angular distance information between the navigation star to the earth and the moon.Considering that the X-ray pulsar signal can continuously provide high-precision navigation information,the X-ray pulsar navigation method is introduced.Through reasonable screening of visible pulsars,extraction of pulse radiation contours,and establishment of pulsar navigation measurement models.In view of the strong nonlinear characteristics of the autonomous celestial navigation system,the Extended Kalman Filter(EKF)and the Cubature Kalman Filter(CKF)are selected to simulate and verify the starlight navigation,pulsar navigation and integrated navigation systems respectively.According to the possible failures of the starlight and pulsar sub-navigation systems in the integrated navigation system in practical applications,multi-condition simulation analysis is carried out to evaluate the performance of the navigation system.Due to the interference of the complicated space environment,the imaging quality of the star sensor is degraded,the navigation star is mismatched,and the observation data has outliers,which affects the accuracy of the n avigation system.To solve this problem,the problem of outlier point modeling is studied,and the Robust Cubature Kalman Filter(RCKF)algorithm is used to achieve high-precision and reliable starlight navigation and integrated navigation system.Finally,through simulation verification,the effectiveness of the method is evaluated.