UKF algorithm of GNSS/CNS/SINS navigation system based on the launching coordinate system

Abstract

The navigation system models of aerospace vehicles and others are usually based on the launch inertial coordinate system (hereinafter referred to as: LICS frame). The commonly adopted navigation mode is the GNSS/CNS/SINS(Global Navigation Satellite System/Celestial Navigation System/Strapdown Inertial Navigation System) multi-fusion mode, and the commonly used fusion algorithm is EKF (Extended Kalman Filter). But EKF will result in decreasing accuracy of navigation due to the linearization processing method adopted by EKF. To overcome the shortcomings of the EKF, an IUKF (improved Unscented Kalman Filter) was designed. In IUKF, each navigation parameter and the errors of navigation sensors were taken as the state vector, while the outputs of GNSS and CNS were directly used as the measurement vector. When EKF and IUKF are used in GNSS/CNS/SINS navigation system, the experimental results show that EKF has stable performance, while the improved UKF algorithm can provide higher-precision navigation parameters.