An In-Motion Alignment Method for Laser Doppler Velocimeter-Aided Strapdown Inertial Navigation System

Abstract

AbstractWith ultra-high velocity measurement accuracy, laser Doppler velocimeter (LDV) is promising to replace the odometer to provide vehicle velocity information in the process of land integrated navigation. This paper investigates the in-motion initial alignment for the LDV-aided strapdown inertial navigation system (SINS). Aiming at the problem that the uncertainty noise in the in-motion alignment process will slow down the alignment speed, a robust square-root unscented quaternion estimator (RSRUSQUE) method is proposed in this paper. The RSRUSQUE method improves the defects of the traditional unscented quaternion estimator (USQUE) method, such as poor noise resistance, slow convergence speed under large misalignment angle and easy to lead to the non-positive definite covariance matrix. This will help to estimate and compensate attitude errors while estimating attitude so as to improve the accuracy of the process model and measurement model, and finally improve the accuracy of attitude estimation. The performance of the proposed scheme 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.