多源信息融合是提高自主水下航行器（AUV）长航程定位精度的常用手段，但复杂海况下DVL偏置与失锁以及惯导误差等会破坏多源相关结构并导致融合解失真甚至发散。为了解决这个问题，本文提出一种基于数据驱动D-S证据理论与混合Copula的自适应卡尔曼融合算法，该方法通过构建多源观测相关性模型以动态量化各信息源的可靠性与冲突，自适应抑制异常信息并联合调整卡尔曼过程与观测噪声协方差，从而提升状态估计精度与鲁棒性。仿真和湖试结果表明，在存在异常信息的情况下能够保持稳定高精度定位，相较于单DVL、INS、Sage-Husa与固定Q/R的卡尔曼算法，本文算法在横向误差、终端误差等都有着显著降低。
Multi-sensor information fusion is a common approach to improving long-range navigation accuracy of autonomous underwater vehicles (AUVs). However, in complex marine environments, DVL bias and outages as well as inertial navigation drift can destroy inter-sensor correlation structures, leading to distorted fusion results or even divergence. To address this issue, this paper proposes an adaptive Kalman fusion algorithm based on data-driven Dempster-Shafer (D-S) evidence theory and hybrid copula modeling. The method constructs a dynamic correlation model among heterogeneous observations to quantitatively evaluate the reliability and conflicts of different information sources, thereby adaptively suppressing abnormal measurements and jointly adjusting the process and measurement noise covariance of the Kalman filter. Simulation and field lake-trial results demonstrate that the proposed algorithm maintains stable and high-accuracy navigation performance in the presence of measurement anomalies, and achieves significant reductions in cross-track error, terminal error, and other metrics compared with single-sensor DVL/INS, Sage–Husa, and fixed-Q/R Kalman filtering methods.
2026,48(6): 100-108 收稿日期：2025-12-2
DOI：10.3404/j.issn.1672-7649.2026.06.014
分类号：U674.941
基金项目：广东省教育厅重点领域专项（2023ZDZX3004）；海洋防务创新基金项目（JJ-2023-715-01）
作者简介：李佳韵(2001-),男,硕士研究生,研究方向为水下机器人多源信息融合定位
参考文献：
[1] 秦洪德, 孙延超. AUV关键技术与发展趋势[J]. 舰船科学技术, 2020, 42(12): 25-28.
QIN H D, SUN Y C Analysis of the status and development of foreign AUV. Ship Science and Technology, 2020, 42(12): 25–28.
[2] 梁益丰, 许江宁, 吴苗, 等. AUV导航技术概述[J]. 舰船科学技术, 2020, 42(15): 152-156+171
LIANG Y F, XU J N, WU M, et al. Overview of AUV navigation technology[J]. Ship Science and Technology, 2020, 42(15): 152-156+171
[3] LI J, GU M, ZHU T, et al. Research on error correction technology in underwater SINS/DVL integrated positioning and navigation[J]. Sensors, 2023, 23(10): 4700
[4] WU Y, TA X, XIAO R, et al. Survey of underwater robot positioning navigation[J]. Applied Ocean Research, 2019, 90: 101845
[5] ZHANG B, JI D, LIU S, et al. Autonomous underwater vehicle navigation: A review[J]. Ocean Engineering, 2023, 273: 113861
[6] SHAUKAT N, MOINUDDIN M, OTERo P. Underwater vehicle positioning by correntropy-based fuzzy multi-sensor fusion[J]. Sensors, 2021, 21(18): 6165
[7] 李冀永, 钟荣兴, 徐雪峰, 等. AUV导航-规划-控制技术研究[J]. 舰船科学技术, 2023, 45(12): 51-56
LI J Y, ZHONG R X, XU X F, et al. Review of navigation, planning and control technology of AUVs[J]. Ship Science and Technology, 2023, 45(12): 51-56
[8] KRAUSS S T, STILWELL D J. Unscented Kalman filtering on manifolds for AUV navigation—experimental results[C]//Proceedings of OCEANS 2022. Hampton Roads: IEEE, 2022.
[9] KALMAN R E. Contributions to the theory of optimal control[J]. Bol. Soc. Mat. Mexicana, 1960, 5(2): 102-119
[10] ZHANG K, LI H, CAO S, et al. Trusted multi-source information fusion for fault diagnosis of electromechanical system with modified graph convolution network[J]. Adranced Engineering Informatics, 2023, 57: 102088
[11] LIU Z. An effective conflict management method based on belief similarity measure and entropy for multi-sensor data fusion[J]. Artificial Intelligence Review, 2023, 56(12): 15495-15522
[12] 张晓林, 汪俊, 严天宏, 等. 基于改进CKF算法的AUV组合导航系统研究[J]. 舰船科学技术, 2025, 47(5): 37-42
ZHANG X L, WANG J, YAN T H, et al. Research on AUV integrated navigation system based on improved CKF algorithm[J]. Ship Science and Technology, 2025, 47(5): 37-42
[13] DU X, PANG X, GUAN F, et al. A novel multi-source navigation algorithm based on factor graph in complex underwater environments of polar regions[J]. Ocean Engineering, 2024, 301: 117516
[14] HUANG J, LI H, LIU Z, et al. GNSS-aided installation error compensation for DVL/INS integrated navigation system using error-state kalman filter[J]. Measurement, 2025, 242: 116224
[15] LIU Z, DONG C, GUO Z, et al. Multi-source domain adaptation fault diagnosis for AUV thrusters based on dynamic weighted learning batch spectral penalization network with dynamic transfer[J]. Ocean Engineering, 2025, 332: 121449
[16] MA H, MU X, HE B. Adaptive navigation algorithm with deep learning for autonomous underwater vehicle[J]. Sensors, 2021, 21(19): 6406
[17] MA X, WEI Z, LIU W, et al. Event-Triggered state filter estimation for INS/DVL integrated navigation with correlated noise and outliers[J]. Sensors, 2025, 25(5): 1545
[18] LI D, XU J, HE H, et al. An underwater integrated navigation algorithm to deal with DVL malfunctions based on deep learning[J]. IEEE Access, 2021, 9: 82010-82020
[19] SUN B, ZHANG Z, QIAO D, et al. An improved innovation adaptive Kalman filter for integrated INS/GPS navigation[J]. Sustainability, 2022, 14(18): 11230
[20] ZHU T, LI J, DUAN K, et al. Study on the robust filter method of SINS/DVL integrated navigation systems in a complex underwater environment[J]. Sensors, 2024, 24(20): 6596
[21] WANG Q, LIU J, JIANG J, et al. Application of improved fault detection and robust adaptive algorithm in GNSS/INS integrated navigation[J]. Remote Sensors, 2025, 17(5): 804
[22] CHOWDHURY M I, PHUNG Q V, AHMED I, et al. Next-generation underwater localization: artificial intelligence-based and energy-aware approaches[J]. Applied Ocean Research, 2025, 165: 104842
[23] IGNATIOUS H A, EL-SAYED H, KULKARNi P. Multilevel data and decision fusion using heterogeneous sensory data for autonomous vehicles[J]. Remote Sensors, 2023, 15(9): 2256
[24] SKLAR A. Fonctions de répartition à n dimensions et leurs marges[J]. Publ. Inst. Statist. Univ. Paris, 1959, 8: 229-231
[25] 季峰, 蔡兴国, 王俊. 基于混合Copula函数的风电功率相关性分析[J]. 电力系统自动化, 2014, 38(2): 1–5.
JI F, CAI X G, WANG J. Wind power correlation analysis based on hybrid Copula[J]. Automation of Electric Power Systems, 2014, 38(2): 1–5, 32.
