A Robust In-Motion Alignment Method for OD-Aided SINS Based on Magnitude Matching

Initial alignment is one of the key technologies of strapdown inertial navigation system (SINS), which will directly affect the accuracy of subsequent navigation. In this paper, the odometer (OD) is used to implement the in-motion alignment process. The OD measurement often produces outliers due to faults or disturbances, which will affect the accuracy of alignment and positioning. Thus, we propose a robust in-motion alignment method based on magnitude matching. First, by using the velocity provided by OD and the specific force and angular velocity provided by SINS, an observation vector and a reference vector in the inertial frame are constructed. Then, an OD outlier detection method is designed based on the magnitude matching of the mentioned vectors. When an outlier is detected, we predict the velocity at the current moment based on historical velocity data and replace the outlier. Finally, the simulation is constructed to demonstrate the effectiveness of the proposed method. The alignment results show that this method can achieve reasonable attitude results, the interference of outliers in OD measurement can be greatly restrained.