Real-time Kinematic Positioning Algorithm in Graphical State Space

Sudan Yan, Shaolin Lü, Gang Liu, Yiding Zhan, Jianan Lou, Rong Zhang

Abstract: This paper concerns the real-time kinematic positioning algorithm in graphical state space. Different from the traditional Kalman filter and the sliding window filter, in our method, graphical state space model which is linearized by Extended Kalman filter is solved by factor graph optimization. Constant variables, such as double-difference ambiguity in a sliding window, will be added into the equation as constants in the graphical state space model, not as time-series variables. Degree of nonlinearity is proposed to analyze the intrinsic properties of traditional discrete-time state space model and graphical state space model. It is shown that sometimes the degree of nonlinearity of the graphical state space model is lower than that of the traditional discrete-time state space model. The potential of the new method is verified by the experiment results.
Published in: Proceedings of the 2023 International Technical Meeting of The Institute of Navigation
January 24 - 26, 2023
Hyatt Regency Long Beach
Long Beach, California
Pages: 637 - 648
Cite this article: Yan, Sudan, Lü, Shaolin, Liu, Gang, Zhan, Yiding, Lou, Jianan, Zhang, Rong, "Real-time Kinematic Positioning Algorithm in Graphical State Space," Proceedings of the 2023 International Technical Meeting of The Institute of Navigation, Long Beach, California, January 2023, pp. 637-648. https://doi.org/10.33012/2023.18676
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In