The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume XLVI-3/W1-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLVI-3/W1-2022, 279–286, 2022
https://doi.org/10.5194/isprs-archives-XLVI-3-W1-2022-279-2022
Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., XLVI-3/W1-2022, 279–286, 2022
https://doi.org/10.5194/isprs-archives-XLVI-3-W1-2022-279-2022
 
22 Apr 2022
22 Apr 2022

SEAFLOOR POSITIONING MODEL FOR SIMULTANEOUS ESTIMATION OF SOUND VELOCITY

S. Zhang1, T. Xu2, and X. Qin3 S. Zhang et al.
  • 1College of Geology Engineering and Geomatics, Chang’an University, Xi’an 710054, Shanxi, China
  • 2Institute of Space Science, Shandong University, Weihai 264209, Shandong, China
  • 3Xi’an Research Institute of Surveying and Mapping, Xi’an 710054, Shanxi, China

Keywords: GNSS-A, Seafloor Geodesy, Positioning, Sound velocity, Temporal and spatial variation, B-Spline

Abstract. The spatial and temporal change of sound speed has a significant impact on the accuracy of GNSS-A underwater positioning. However, it is hard to collect enough data to cover all the spatial and temporal changes of sound velocity. We built an acoustic ranging model that included the sound velocity component, and then we looked at how the model could be used for three-dimensional seabed location and undersea crust monitoring. Meanwhile, B-spline curves are used to construct a sound velocity model that encompasses temporal variation as well as two-dimensional spatial gradients. The method was utilized to assess the simulated data and the in-situ data collected in July 2019, respectively. Simulation results show that the root mean square of the horizontal coordinates solved by the ranging model is less than 10 cm, and the model may meet the demands of subsea crust monitoring under certain situations. In the real data experiment, the square root of variance of the coordinate was better than 10 cm. The sound velocity model findings indicated that the variance of sound velocity in the experimental marine region was less than 1m/s, with clear daily patterns and short-period variations. The algorithm does not need the sound velocity profile information and can save the measurement time of the sound velocity profile.