基于状态量扩维的旋转式捷联惯导系统精对准方法

Initial alignment is a key technology of rotary strapdown inertial navigation system (SINS). The existing 10D model of traditional rotary SINS is mostly used for the fine alignment, which cannot meet the requirements of navigation accuracy. To solve this problem, a fine alignment method for rotary SINS based on augmented state is proposed. First, scale factor errors and installation errors are extended to state variables, and a 28D fine alignment model is established. Second, the observability of 28D model during rotation is analyzed. Third, a 13D alignment model is designed according to analysis results. Finally, Kalman filter is used to achieve the fine alignment of rotary SINS. The simulation results show that the proposed method can effectively improve the attitude alignment accuracy and estimate more gyroscope error terms compared with the traditional initial alignment method.