Robust Vector Tracking Loop Using Moving Horizon Estimation

Vector tracking is an approach to simultaneously process GNSS signal tracking and position, velocity, time (PVT) estimation, commonly based on an extended Kalman Filter (EKF). In vector tracking, the parameters of locally generated signals are determined by the estimated receiver PVT and satellite information, thus the channels are aided by each other. The inter-channel aiding gives vector tracking several benefits when processing signals of low carrier-to-noise power density ratio (C/No) and high dynamics. However, vector tracking has one significant drawback. That is the presence of bad measurement data (or fault) in one channel will affect the other channels, and in the worst case, could lead to loss of lock in all channels. To deal with this problem, we propose a Moving Horizon Estimation (MHE) technique for vector tracking. The basic idea of MHE is to reformulate the estimation problem as a quadratic programming (QP) problem within a moving, fix-sized estimation window. A main advantage of MHE is that it naturally allows the incorporation of inequality constraints on the state vector and disturbance. When applied to vector tracking, MHE can constrain the effect from each channel within a priori defined range, thus mitigate the effect from the faulty channels. Another advantage of MHE is that it is less sensitive to tuning parameters, in other words, MHE is more robust to environment change comparing with EKF. Simulation results are presented to demonstrate the improved performance of vector tracking using MHE in scenarios of signal attenuation and ionosphere scintillation.