Precise positioning in complex environments with GNSS-RTK and weighted linear regression UWB observations

ABSTRACT

In complex urban, port, and factory environments, GNSS and UWB encounter significant challenges from signal blockage and multipath effects. This study proposes a GNSS-RTK/WLR-UWB tight integration positioning method, utilizing Weighted Linear Regression to fit missing UWB epochs and introducing dynamic weight adjustment. The track cart and pedestrian experiment demonstrates that the single-epoch positioning accuracy is improved by 53.47%, 65.94%, and 28.03% in E, N and U compared to RTK. The filtering positioning achieves accuracy better than 1 cm in E and N, with a high ambiguity resolution success rate exceeding 99.40%, and reduces time-to-first-fix from 6-13 s to 3 s.

KEYWORDS:

• Global navigation satellite system (GNSS)
• real-time kinematic (RTK)
• ultra wide band (UWB)
• tight integration positioning
• ambiguity resolution

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [42022025]; The China National Railway Group Co., Ltd. Major Project [N2023X003].