GPS Vector Tracking Loop Enhancement Using a Robust Cubature Kalman Filter

Global Positioning System (GPS) receiver is a low-cost apparatus for acquiring precise positioning, velocity and timing information. Basically, it has two different architectures, termed as "Scalar-Tracking' and "Vector-Tracking". Among these two architectures, Vector Tracking Loop (VTL) has been demonstrated better performance than Scalar Tracking Loop (STL). VTL processes the signal tracking together using a center navigation filter, while the STL operates the signal tracking independently. Mutual aiding between channels allows the VTL better performance than STL. However, the navigation filter of the VTL is usually a Kalman filter, and the nonlinearity and disturbance of the measurements will affect the VTL operating. In this paper, a cubature Kalman filter (CKF) was employed to solve the nonlinear problem, and a robust M estimation scheme was utilized for overcoming the abnormal measurements. A 3D dynamic trajectory was designed for testing and evaluating the adaptive robust CKF (AR-CKF) performance in VTL.