A Multi-Vehicle Cooperative Positioning Method Based on Factor Graph Optimization Using the Error Information of Cooperators

The positioning technology based on the Global Navigation Satellite System (GNSS) often encounters significant challenges in complex environments, such as buildings, urban canyons, and tunnels, due to multipath effects and non-line-of-sight (NLOS) signals, making it difficult to meet the requirements of high-precision vehicle positioning. Cooperative positioning (CP) methods can enhance the accuracy and reliability of vehicular positioning in GNSS-degraded environments. However, traditional CP methods often overlook the positioning error information of cooperators, leading to a detrimental impact on the overall accuracy of CP systems. In this paper, we propose a novel CP method that integrates real-time error information of cooperators to address this limitation. To improve the reliability and accuracy of the CP system, we utilize Factor Graph Optimization (FGO) to fuse the available information. Unlike traditional algorithms that rely solely on current epoch observations, FGO algorithms employ global optimization algorithms to overcome the limitations. The key innovation of our proposed CP method lies in the real-time evaluation of vehicle accuracy and the utilization of this information to constrain the uncertainty of the cooperators in challenging urban environments. Numerical results and comparisons validate the feasibility and superiority of the proposed method.