ISSN 2096-4498

   CN 44-1745/U

二维码

Tunnel Construction ›› 2026, Vol. 46 ›› Issue (10): 2240-2249.DOI: 10.3973/j.issn.2096-4498.2026.10.014

Previous Articles     Next Articles

Single-Cross-Section Positioning of Immersed-Tunnel Guide Rods Based on Omnidirectional Sonar

SUO Xuhong1, 2, YAN Haowen1, LI Zhenhong1, 3, *, CHE Tao1, ZHANG Dejin4,  GUAN Minglei5, LU Kebang6, AN Mingzan4   

  1. (1. Faculty of Geomatics, Lanzhou Jiaotong University, Lanzhou 730070, Gansu, China; 2. CCCC First Harbor Engineering Company Ltd., Tianjin 300461, China; 3. School of Geological Engineering and Geomatics, Chang’an University, Xi’an 710054, Shaanxi, China; 4. School of Architecture & Urban Planning, Shenzhen University, Shenzhen 518060, Guangdong, China; 5. School of Artificial Intelligence, Shenzhen Polytechnic University, Shenzhen 518055, Guangdong, China; 6. College of Civil Transportation Engineering, Shenzhen University, Shenzhen 518060, Guangdong, China)
  • Online:2026-10-20 Published:2026-10-20

Abstract: An omnidirectional sonar method was developed to continuously and quantitatively determine the local positions of guide rods during immersed-tunnel element connection. The slot center, sonar source center, scan direction, and theoretical rod-section center were jointly calibrated within a unified construction coordinate frame. Range-bearing observations were converted into cross-sectional coordinates. Position-prior, spatial-continuity, and known-radius constraints were applied to extract wall echoes, and Huber robust circle fitting was used to estimate the cross-sectional center. Echo count, arc coverage, and fitting residuals were used to assess the validity of the estimated transverse and elevation deviations. Validation included pool tests of cylindrical targets at distances of 0.8-2.8 m, nine Monte Carlo distance-arc scenarios with 1 000 repetitions each, comparisons of fitting methods, sensitivity analyses of prior bias and scan tilt, and field observations during the installation of element E3 of the Tanzhou Tunnel. The results were as follows: (1) Pool tests yielded continuously identifiable echoes from cylindrical-targets, enabling local arc extraction. (2) Across the nine simulated scenarios, the root mean square error (ERMS) of the estimated center ranged from 2.76 to 6.33 mm, while the 95th-percentile error ranged from 4.90 to 10.99 mm. The known-radius constraint reduced the overall ERMS by approximately 94.9% relative to free-radius least squares, and Huber weighting yielded a further 15.2% reduction. (3) Of the 40 field records, 32 passed quality control, corresponding to an acceptance rate of 80.0%. Among the valid records, 25 showed absolute transverse differences of no more than 15 mm from measurement-tower estimates, and 26 showed differences of no more than 18 mm. These findings support the use of the method for auxiliary verification of local guide-rod positions. Field comparisons demonstrate consistency within the same datum rather than absolute accuracy against independent ground truth. Applications to long immersed tunnels require error budgets tailored to the actual control network.

Key words: immersed tunnel, guide-rod positioning, omnidirectional sonar, cross-sectional positioning, Huber robust circle fitting, quality control