.png){kind=icon}
.png){kind=icon}
Return
An Advanced Resilient Ambiguity Resolution Strategy Accounting for Non-Gaussian Distribution for Urban GNSS Real-time Kinematic Positioning
.png){kind=icon}
PRE
AI
W
X
Y
H
X
J
DOI：10.1016/j.asr.2026.05.021.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Correct ambiguity resolution is crucial for high-precision Global Navigation Satellite System (GNSS) Real-time Kinematic (RTK) positioning. While the existing batch Resilient Ambiguity Resolution (RAR) strategy performs well in complex environments, it overlooks the distributional assumptions of observation errors, potentially degrading performance. We propose an advanced RAR strategy that integrates heavy-tailed distributions, especially the multivariate t-distribution, to enhance error representation. Meanwhile, this strategy considers differences in fractional cycle biases and ambiguity candidate counts across distributions, thereby implementing appropriate thresholds. Experimental results demonstrate several key findings. In medium urban areas, the advanced strategy utilizing the multivariate t-distribution (ARAR-t) achieves comparable accuracy to the traditional strategy (TRAR), while significantly improving the east, north, and up components by 51.4%, 37.5%, and 60.6%, respectively, compared to the Integer Least Squares (ILS) solution. In deep urban environments, vertical positioning accuracy exhibits remarkable enhancements, with average improvements of 53.4% over ILS and 12.8% over TRAR. Notably, the vertical error of the worst dataset is reduced from 0.656 m (ILS) to 0.510 m (TRAR) and further to 0.417 m (ARAR-t). Additionally, the Laplace distribution shows comparable accuracy to ARAR-t in most scenarios, whereas the Minmax distribution proves unsuitable. In conclusion, the advanced ARAR-t strategy offers a more reliable solution for urban GNSS RTK positioning, demonstrating improved adaptability to real-time environmental variations.
Keywords:
Resilient Ambiguity Resolution
Non-Gaussian Distribution
Urban GNSS RTK
Multivariate t-distribution
Integer Least Squares
Journal
IF:
2.8
Papers:
1.3K
Citations:
2.0W
Organization
Cited Papers