In-Motion Initial Alignment Method for LDV-Aided SINS Based on Robust Unscented Quaternion Filter

Abstract

With the advantages of high velocity measurement accuracy, good spatial resolution and fast dynamic response, the laser Doppler velocimeter (LDV) is expected to replace the odometer and Doppler Velocity Log to be combined with a strapdown inertial navigation system (SINS) to form a higher precision integrated navigation system. Currently, in-motion initial alignment of SINS is still a challenge. In this paper, the high precision velocity provided by LDV to aid SINS in-motion alignment. Considering that some approximation used in the alignment model and the unknown noise parameters during the filter process, an unscented quaternion H-infinite estimator (USQUHE) is proposed in this paper. Because USQUHE combines the advantages of H-infinite filter and unscented quaternion estimator, USQUHE has satisfactory robustness when processing nonlinear models. The performance of the proposed method is verified by a vehicle field test. The results show that the proposed method has higher alignment accuracy, faster convergence speed and stronger robustness than other compared methods.