Scholarlink

<p>The spatiotemporal variations of sound speed, particularly the drastic variation in depth, significantly affect seafloor geodetic positioning precision. For this reason, the global navigation satellite system-acoustic (GNSS-A) positioning technology typically uses <i>in-situ</i> sound speed profiles (SSPs) and considers the impact of these variations at the data post-processing stage. However, <i>in-situ</i> SSP measurement is costly and somewhat hinders the timeliness of seafloor geodetic monitoring. We generalize the bilinear SSP (BL-SSP) to be a piecewise-linear SSP, whose model parameters are estimated from GNSS-A observations. In addition, we construct a set of constraints based on <i>a priori</i> marine environment observation to stabilize SSP inversion and propose an algorithm to recursively conduct the inversion, e.g., the trilinear SSP (TL-SSP) inversion is initialized using the BL-SSP inversion result. The proposed model is verified by long-term GNSS-A seafloor geodetic observations. It shows that the root mean square error (RMSE) of the TL-SSP inversion result is 10.87 m/s, compared to 11.08 m/s for the traditional BL-SSP, with significant improvements observed in shallow and middle water layers. Furthermore, when replacing the <i>in-situ</i> SSP with the inverted SSP for precise seafloor geodetic positioning and incorporating the acoustic delay parameters, the TL-SSP-based positioning demonstrates higher accuracy than the BL-SSP-based approach. Relative to the positioning result based on the <i>in-situ</i> SSP, the mean bias, standard deviation and RMSE of the horizontal positioning error are better than 0.003 m, 0.005 m, and 0.006 m, respectively, while those of the vertical positioning error are better than 0.03 m, 0.04 m, and 0.04 m, respectively. Compared with BL-SSP, TL-SSP can achieve a positioning error reduction along the E-direction, N-direction, and U-direction by 16.7%, 15.0%, and 5.5%, respectively.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

GNSS-A positioning model with piece-wise linear sound speed profile inversion

  • 期刊论文

  • 发表日期：

    2025年9月1日

查看全文
  • Jixing Zhu,
  • Shuqiang Xue,
  • Baojin Li,
  • Zhen Xiao,
  • Jiachao Bian,
  • Yunhao Fan

摘要

The spatiotemporal variations of sound speed, particularly the drastic variation in depth, significantly affect seafloor geodetic positioning precision. For this reason, the global navigation satellite system-acoustic (GNSS-A) positioning technology typically uses in-situ sound speed profiles (SSPs) and considers the impact of these variations at the data post-processing stage. However, in-situ SSP measurement is costly and somewhat hinders the timeliness of seafloor geodetic monitoring. We generalize the bilinear SSP (BL-SSP) to be a piecewise-linear SSP, whose model parameters are estimated from GNSS-A observations. In addition, we construct a set of constraints based on a priori marine environment observation to stabilize SSP inversion and propose an algorithm to recursively conduct the inversion, e.g., the trilinear SSP (TL-SSP) inversion is initialized using the BL-SSP inversion result. The proposed model is verified by long-term GNSS-A seafloor geodetic observations. It shows that the root mean square error (RMSE) of the TL-SSP inversion result is 10.87 m/s, compared to 11.08 m/s for the traditional BL-SSP, with significant improvements observed in shallow and middle water layers. Furthermore, when replacing the in-situ SSP with the inverted SSP for precise seafloor geodetic positioning and incorporating the acoustic delay parameters, the TL-SSP-based positioning demonstrates higher accuracy than the BL-SSP-based approach. Relative to the positioning result based on the in-situ SSP, the mean bias, standard deviation and RMSE of the horizontal positioning error are better than 0.003 m, 0.005 m, and 0.006 m, respectively, while those of the vertical positioning error are better than 0.03 m, 0.04 m, and 0.04 m, respectively. Compared with BL-SSP, TL-SSP can achieve a positioning error reduction along the E-direction, N-direction, and U-direction by 16.7%, 15.0%, and 5.5%, respectively.