A robust and continuous carrier phase prediction strategy for GNSS/INS deeply coupled systems

• Published in: GPS solutions 2024-10, Vol.28 (4), p.168, Article 168
• Main Authors: Zheng, Qiyuan, Jiang, Jinguang, Yan, Peihui, Wu, Jiaji, Li, Yuyin, Tan, Hongbin, Liu, Jianghua
• Format: Article
• Language: English
• Subjects: Atmospheric Sciences; Automotive Engineering; Earth and Environmental Science; Earth Sciences; Electrical Engineering; Error analysis; Geophysics/Geodesy; Global navigation satellite system; Navigation systems; Performance prediction; Phase error; Review Article; Satellites; Signal processing; Space Exploration and Astronautics; Space Sciences (including Extraterrestrial Physics
• ISSN: 1080-5370; 1521-1886
• DOI: 10.1007/s10291-024-01696-6

The global navigation satellite system’s signals are frequently obstructed in complex environments, and the carrier phase is prone to experiencing large cycle slips; hence, phase prediction becomes necessary. However, the current phase prediction theory presents flaws in the prediction preparation stage. The existing prediction methods merely resort to the error threshold to determine the start time of prediction, which may give rise to significant initial prediction errors. We tackle this problem and its corresponding solution for deeply coupled systems: the multistage threshold discrimination method. This method analyzes the phase error information of the cache in the prediction preparation stage to accurately determine the start time of prediction and minimize the initial prediction error. The performance of the proposed method is evaluated with static and dynamic data. In predicting for 20 s under static conditions and 10 s under dynamic conditions, the predicted pass rates are 92.6% and 52.4%, respectively, 10.4% and 11.0% higher than those of the original method. The average prediction error is reduced by 36.6% and 33.9% under static and dynamic conditions. In the scenario where the signal is interrupted multiple times, the root mean square of positioning error is reduced by 30.2%, 53.0%, and 58.7% in the east, north, and up directions. These results suggest that the proposed method is effective and constitutes a complement to the phase prediction theory.