Real-time Multi-Mode Robust Filtering Method for Rocket Upper Stage

A real-time multi-mode robust filtering method is proposed for the large thrust and slow spin characteristics of the rocket above the flight stage. The rocket upper stage in the sliding section of the existence of slow spin characteristics, which will lead to ground measurement equipment tracking instability, poor measurement data quality. In order to solve this problem, the robustness theory and CKF filter algorithm are combined to improve the system robustness. The rocket upper stage in the track stage, because the thrust is too large lead to poor performance tracking. The method of multi-modal orbit determination based on embedded maneuvering decision method can adjust the state variance matrix at the time of maneuver, which can speed up the correction of the system information and improve the tracking performance of the system. Through a rocket upper stage experimental data analysis, this method can effectively reduce the measurement data of poor quality problem, improve the system tracking performance, has a certain application value for the upper stage tracking trajectory reconstruction.