GPT3 model accuracy and it’s effects on precise point positioning in the east Vietnam sea region
PDF (Tiếng Việt) | Download: 96

Working Languages

How to Cite

Nguyen, N. L., & Pham, A. D. (2021). GPT3 model accuracy and it’s effects on precise point positioning in the east Vietnam sea region. Journal of Geodesy and Cartography, (48), 1–7. https://doi.org/10.54491/jgac.2021.48.242

Abstract

We extract the tropospheric zenith delay (TZD) from 19 days of GNSS data (October 29 to November 16, 2020) located in the peak of the rainy season in the East Vietnam Sea area of two GNSS permanent stations in the Philippines and Vietnam. When compared with the results of the TZD calculation from the GPT3 model, the maximum deviation is up to more than 1dm. This leads to precise point positioning (PPP) when correcting TZD using GPT3 and VMF3 models reduces the success rate of ambiguity resolution and causes large systematic errors in the height component. The positioning errors in the North, East and Up components were achieved (0.005, 0.004, 0.136) m when 24 hour static and (0.018, 0.018, 0.136) m when kinematic processing. Considering the TZD as an unknown to estimate together with the station coordinates, the PPP processing has the success rate of ambiguity resolution being 97% and the positioning error obtained as (0.002, 0.002, 0.010) m when 24h static and (0.015, 0.018, 0.050) m when kinematic processing. The Up component accuracy is improved by 3-13 times compared to using the GPT3 model.
https://doi.org/10.54491/jgac.2021.48.242
PDF (Tiếng Việt) | Download: 96

Downloads

Download data is not yet available.