Acta Geodaetica et Cartographica Sinica ›› 2026, Vol. 55 ›› Issue (3): 515-524.doi: 10.11947/j.AGCS.2026.20250356

• Marine Surveying and Mapping • Previous Articles     Next Articles

Multi-window joint robust estimation for marine acoustic navigation

Jiachao BIAN1(), Shuqiang XUE1(), Shuang ZHAO2, Jixing ZHU1, Jinlai GAO1, Baojin LI2   

  1. 1.State Key Laboratory of Spatial Datum, Chinese Academy of Surveying and Mapping, Beijing 100036, China
    2.China University of Petroleum (East China), Qingdao 266580, China
  • Received:2025-09-01 Revised:2026-03-05 Online:2026-04-16 Published:2026-04-16
  • Contact: Shuqiang XUE E-mail:bjc0539@163.com;xuesq@casm.ac.cn
  • About author:BIAN Jiachao (2000—), male, master, major in ocean acoustic navigation. E-mail: bjc0539@163.com
  • Supported by:
    The National Key Research and Development Program(2024YFB3909702);The National Natural Science Foundation of China(42474014);Laoshan Laboratory Project(LSKJ202205100)

Abstract:

Marine acoustic navigation typically employs active sonar to obtain the round-trip signal propagation time between the carrier and the navigation beacon. However, it cannot simultaneously acquire multi-beacon acoustic observations, and it is difficult to implement acoustic observation quality control solely relying on single-epoch acoustic observations. To address this issue, this paper proposes a windowed robust least squares estimation algorithm. By implementing a multi-window joint robustness strategy, the algorithm dynamically constructs robust equivalent weights using observation quality information within historical windows during the window sliding process. Specifically, the initial weights for observations in the new window are determined by taking the mean of the robust equivalent weights across multiple historical windows, and the quality of newly added observations within the window is evaluated using the carrier trajectory model prediction information. Experimental results show that: ① under Huber, IGG Ⅱ, and IGG Ⅲ robustness strategies, the proposed algorithm can effectively resist the impact of gross errors, especially significantly enhancing the robustness performance of lever observations at the edge of the window; ② the proposed algorithm can significantly improve the accuracy and reliability of navigation and positioning results, resulting in smoother and more stable robust navigation trajectory estimation.

Key words: acoustic navigation, sliding-window, outlier detection, robust estimation, IGG Ⅲ

CLC Number: