A Robust Cubature Kalman Filter for GPS Vector Tracking Loop

Boya Zhang, Shuai Chen, Xiaohan Zhu

Vector Tracking Loop (VTL) is a recently proposed method to enhance GPS receiver performance. In VTL, a center navigation filter usually a Kalman filter (KF) is utilized to estimate navigation solutions and complete signal tracking together. Thus, all channels are processed together and mutual aiding can be obtained. Compared with Scalar Tracking Loop (STL), in which signal tracking is operated independently, researches have demonstrated that VTL performs better signal tracking performance. However, the nonlinear problems or the measurement outliers might affect the navigation filter and hinder VTL performance. This paper investigates applying robust adaptive cubature Kalman filter (AR-CKF) to VTL navigation filter. Robust M estimation is employed to resistant the measurement outliers and an adaptive factor is utilized to address the dynamic disturbance errors. A 3D dynamic trajectory is generated to test the AR-CKF based VTL. Simulations are implemented in a VTL software receiver, the results from comparing a common Kalman Filter with AR-CKF, which demonstrates that the employed AR-CKF improves VTL stability and accuracy.