Abstract
With the growing demand for urban positioning and navigation, Global Navigation Satellite System services are increasingly essential in smart-city applications. However, signal occlusion caused by dense buildings often leads to non-line-of-sight (NLOS) reception; existing methods typically adopt fixed exclusion or reweighting strategies to handle NLOS signals, which can either weaken satellite geometry or allow residual NLOS errors to propagate into the solution. To address this issue, we propose a geometry-adaptive (GA) sliding-window factor-graph optimization framework which performs GA NLOS handling and jointly optimizes pseudorange and time-differenced carrier-phase factors in a sliding-window factor graph. A vehicle field experiment was conducted to validate the effectiveness of the proposed method. Results indicate that the proposed method achieves positioning RMSEs of 1.83 m and 2.32 m in the horizontal and vertical directions, corresponding to reductions of 61.8% and 59.2%, respectively, compared with conventional methods. Particularly, in the deep urban scenario, the proposed method improves the 3D positioning accuracy by up to 72.1%.
Funding Information
  • Postdoctoral Project of Hubei Province (2024HBBHCXB060)
  • National Natural Science Foundation of China (42174030)
  • Scientific Research Innovation Capability Support Project for Young Faculty (SRICSPYF-ZY2025007)
  • Major Program (JD) of Hubei Province (2023BAA025)
  • State Key Laboratory of Geodesy and Earth’s Dynamics, Innovation Academy for Precision Measurement Science and Technology, Chinese Academy of Sciences (SKLPG2025-5-1)
  • Natural Science Foundation of Wuhan (2024040701010029)

This publication has 40 references indexed in Scilit: