A new method of initial alignment and self-calibration based on dual-axis rotating strapdown inertial navigation system

Initial alignment is one of the key technologies of the inertial navigation. The initial alignment precision affects the navigation result. Researches show that the result of initial alignment depends on the observability of the system state variables. So it's necessary to enhance the system states observability to improve the accuracy and speed. A positioning method based on dual-axis rotating strapdown initial navigation system (SINS) is introduced in this paper which enhances the system states observability and improves the observability degree by changing the system error model matrix periodically. The PWCS (Piece-Wise Constant System) and SVD (Singular Value Decomposition) are used to analyse the observability. The results show that the system observability is improved; the misalignment angles and the gyroscope constant drifts are estimated in 1 hour, and accelerometers bias are estimated in 0.8 hour with the new dual-axis rotating inertial alignment and self-calibration method.