GPS/INS Integrated Navigation Based on Unscented Kalman Filter

Abstract

[Purpose] In GPS/INS integrated navigation, which is widely used in high precision of the real-time navigation, the Extended Kalman Filter (EKF) has become one of the most widely used algorithms. Unfortunately, the EKF is based on a sub-optimal implementation of the recursive Bayesian estimation framework applied to Gaussian random variables. This can seriously affect the accuracy or even lead to divergence of the system. In order to improve the accuracy, we apply the Unscented Transformation to GPS/INS integrated navigation. [Method] This paper optimizes GPS/INS integrated navigation by applying the Unscented Kalman Filter (UKF) algorithm which is based on the Unscented Transformation. [Results] The experimental results show that the UKF has an error reduction of over 10% in every estimator relative to the EKF. [Conclusions] Consequently, the UKF is an effective algorithm to improve the accuracy of GPS/INS integrated navigation.