Adaptive Stochastic Model based on LS-VCE for Observations in Kinematic Precise Point Positioning

Abstract

For kinematic precise point positioning, the stochastic model plays an important role in optimizing float trajectories and improving float solution convergence. However, the stochastic model currently used is mainly an empirical function model, which cannot accurately reflect the true error level of observations in complex environments in real time. To address this problem, an adaptive stochastic model based on least-squares variance component estimation (LS-VCE) method is proposed in this paper. The coefficients of the stochastic model are adaptively adjusted by real-time estimation of the unit weight variance factor of the pseudorange and carrier-phase observations. The optimal estimation of the positioning result is achieved by fusing LS-VCE and Kalman filter. The effectiveness of the proposed method is verified by both static and kinematic tests. The results show that the proposed method can significantly improve the positioning accuracy and stability of precise point positioning.