Underwater transponder position calibration method based on triangular partition optimization gauss-newton algorithm
By optimizing the initial value selection of the Gauss-Newton algorithm using the triangular segmentation method and introducing a cost function to handle abnormal information, the problems of improper initial value selection and abnormal measurement information in underwater transponder position estimation are solved, thereby improving the accuracy and stability of underwater positioning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HOHAI UNIV
- Filing Date
- 2022-11-30
- Publication Date
- 2026-06-19
AI Technical Summary
Existing underwater transponder position estimation methods suffer from local optima or divergence problems due to improper initial value selection in underwater environments, as well as large estimation errors when measurement information is abnormal, which affect the positioning accuracy of ultra-short baseline systems.
The initial value selection of the Gauss-Newton algorithm is optimized by using the triangular partitioning method, and a cost function is introduced for weighted processing. The optimal initial value is selected by using the triangular partitioning method, and the anomaly handling of measurement information is performed by combining the cost function, thereby improving the stability and accuracy of the algorithm.
This method improves the convergence speed of the Gauss-Newton method and the robustness of transponder position estimation, solves the problem of poor positioning accuracy caused by improper initial value selection and abnormal measurement information, and achieves fast and accurate underwater transponder position calibration.
Smart Images

Figure CN116087926B_ABST
Abstract
Citation Information
Patent Citations
Acoustic positioning system of shallow sea wave detector
CN101793965A
Ultra-short baseline five-element stereoscopic space array and underwater acoustic localization method
CN106199519A