测绘学报 ›› 2025, Vol. 54 ›› Issue (10): 1841-1851.doi: 10.11947/j.AGCS.2025.20240469

• 海洋测绘 • 上一篇    下一篇

水下地形匹配定位抗差粒子滤波算法

李根1(), 柴洪洲1(), 靳凯迪2, 展昭1   

  1. 1.信息工程大学地理空间信息学院,河南 郑州 450001
    2.军事科学院战争研究院,北京 100091
  • 收稿日期:2024-11-19 修回日期:2025-09-05 出版日期:2025-11-14 发布日期:2025-11-14
  • 通讯作者: 柴洪洲 E-mail:a1145216790@163.com;chaihz1969@163.com
  • 作者简介:李根(2002—),男,博士生,研究方向为水下UUV自主定位。E-mail:a1145216790@163.com
  • 基金资助:
    国家自然科学基金(42404010);智能空间信息国家级重点实验室基金(SYS-ZX01-2024-01)

Underwater terrain matching method based on robust particle filter

Gen LI1(), Hongzhou CHAI1(), Kaidi JIN2, Zhao ZHAN1   

  1. 1.Institute of Surveying and Mapping, Information Engineering University, Zhengzhou 450001, China
    2.Institute of War Studies, Academy of Military Science, Beijing 100091, China
  • Received:2024-11-19 Revised:2025-09-05 Online:2025-11-14 Published:2025-11-14
  • Contact: Hongzhou CHAI E-mail:a1145216790@163.com;chaihz1969@163.com
  • About author:LI Gen (2002—), male, PhD candidate, majors in UUV autonomous navigation. E-mail: a1145216790@163.com
  • Supported by:
    The National Natural Science Foundation of China(42404010);The National Key Laboratory of Intelligent Spatial Information(SYS-ZX01-2024-01)

摘要:

地形匹配辅助导航(TAN)系统可以修正水下无人航行器(UUV)的位置误差,实现UUV水下绝对定位。针对单波束测深值易受粗差影响,导致基于粒子滤波(PF)的TAN系统易发散的问题,本文提出了一种基于抗差粒子滤波的水下地形匹配定位方法。通过分析测深粗差影响UUV水下地形匹配定位的作用机理,在PF地形匹配中引入抗差估计方法,使用IGG Ⅲ函数设置抗差因子,动态调节粗差观测值对后验状态参数的贡献。蒙特卡洛仿真试验表明,导航结束时,相较于标准PF算法,本文算法的精度与稳定性分别提高了12.20%与58.81%。引入抗差因子后,本文算法面对不同类型粗差时,均表现出了良好的精度与稳定性。

关键词: 水下无人航行器, 捷联惯性导航系统, 地形匹配定位, 粒子滤波, 抗差估计

Abstract:

Terrain-aided navigation (TAN) systems are capable of correcting the position errors of unmanned underwater vehicle (UUV), enabling absolute positioning underwater. This paper addresses the issue that single-beam sounding values are susceptible to gross errors, which can degrade the positioning accuracy of TAN systems based on particle filters (PF). To address this, a robust particle filter-based underwater terrain-matching localization method is proposed. By analyzing the mechanism by which gross errors in sounding affect UUV underwater terrain matching, a robust estimation method is introduced into the PF terrain matching, utilizing the IGG Ⅲ function to set robust factors that dynamically adjust the contribution of gross error observations to the posterior state parameters. Monte Carlo simulation experiments show that at the end of navigation, compared to the standard PF algorithm, the accuracy and stability of the proposed algorithm are improved by 12.20% and 58.81%, respectively. After introducing the robust factor, the proposed algorithm demonstrated better accuracy and robustness when facing different types of gross errors.

Key words: unmanned underwater vehicle, strapdown inertial navigation system, terrain-aided navigation, particle filter, robust estimation

中图分类号: