GNSS/INS Positioning in Dense Urban Environment with Adaptive Choice of Process Noise Covariance Based on Satellite Geometry

Yoji Takayama, Takateru Urakubo, Hisashi Tamaki

Peer Reviewed

Abstract: Satellite visibility from a GNSS receiver moving in dense urban environments has two difficult properties: lack of visibility and rapid change in visibility. Buildings, walls, and other objects around the receiver block many satellite signals, reducing the number of visible satellites and biasing the geometric distribution of visible satellites. This lack of visibility degrades the positioning accuracy and tends to cause excessive inflation of the estimation error covariance matrix when using the extended Kalman filter. The movement of a receiver causes a rapid change in visibility; for example, a satellite that is hidden behind a building becomes suddenly visible. As pointed out in this paper, even when the number of visible satellites increases through the visibility change, the positioning accuracy obtained from the extended Kalman filter can be degraded due to the excessively inflated covariance matrix. The H-adaptive filter that we have proposed in our previous work can avoid the degradation of accuracy in dense urban environments. By adaptively choosing the process noise covariance matrix to the measurement matrix at each time step, the excessive inflation of the estimation error covariance matrix can be eliminated. In this paper, we apply the H-adaptive filter to GNSS/INS positioning and experimentally demonstrate that positioning accuracy is improved compared to those obtained from two filters, a standard extended Kalman filter, and a fading memory filter. The experimental results show that positioning errors are reduced by 10 [%] to 40 [%] with the proposed filter.
Published in: Proceedings of the 36th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2023)
September 11 - 15, 2023
Hyatt Regency Denver
Denver, Colorado
Pages: 1974 - 1983
Cite this article: Takayama, Yoji, Urakubo, Takateru, Tamaki, Hisashi, "GNSS/INS Positioning in Dense Urban Environment with Adaptive Choice of Process Noise Covariance Based on Satellite Geometry," Proceedings of the 36th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2023), Denver, Colorado, September 2023, pp. 1974-1983. https://doi.org/10.33012/2023.19325
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In