A Hybrid UKF–ST-Graph Transformer Framework for Cooperative Localization in Ship Ad-Hoc Networks

Anfal Ibraheem; Sumaya Hamad

doi:10.35377/saucis...1873223

A Hybrid UKF–ST-Graph Transformer Framework for Cooperative Localization in Ship Ad-Hoc Networks

Abstract

In a sea environment, the localization of ships is very important for traffic management. The Automatic Identification System (AIS) dataset is widely used to get information about ships. This paper proposed a hybrid localization framework for Ship Ad Hoc Networks (SANETs) using the AIS dataset. This study integrates an Unscented Kalman Filter (UKF) with a Spatio-Temporal Graph Transformer (ST-GT) built on SANETs to provide cooperative localization. First, UKF utilizes raw AIS data to get initial position estimates for ships. Then, a SANET based on spatial proximity is constructed to form features such as node degree and local connectivity. These features are used by the ST-GT model to obtain temporal motion patterns and inter-vessel interactions to enhance that initial position. The suggested framework outperforms the standalone UKF in experimental results on real AIS datasets. The localization errors under sparse and noisy environments are decreased by using this framework. The result shows the effectiveness of combining physical motion models with deep learning through SANETs for enhanced maritime localization.

Keywords

Ethical Statement

The authors declare that the scientific and ethical principles were followed, and all the studies that benefited from are stated in the bibliography.

References

J. Wenzhe, T. Haina, and Z. Xudong, “STGDPM: Vessel trajectory prediction with spatio-temporal graph diffusion probabilistic model,” arXiv:2503.08065, Mar. 2025, doi: 10.48550/arXiv.2503.08065. [Online]. Available: https://doi.org/10.48550/arXiv.2503.08065
S. Fossen and T. I. Fossen, “Exogenous Kalman filter (XKF) for visualization and motion prediction of ships using live automatic identification system (AIS) data,” Model. Identif. Control, vol. 39, no. 4, pp. 233–244, 2018, doi: 10.4173/mic.2018.4.1. [Online]. Available: https://doi.org/10.4173/mic.2018.4.1
Z. Xie, E. Tu, X. Fu, G. Yuan, and Y. Han, “AIS data-driven maritime monitoring based on transformer: A comprehensive review,” arXiv:2505.07374, May 2025, doi: 10.48550/arXiv.2505.07374. [Online]. Available: https://doi.org/10.48550/arXiv.2505.07374
X. Peng, B. Zhang, and L. Rong, “A robust unscented Kalman filter and its application in estimating dynamic positioning ship motion states,” J. Mar. Sci. Technol., vol. 24, no. 4, pp. 1265–1279, Dec. 2019, doi: 10.1007/s00773-019-00624-5. [Online]. Available: https://doi.org/10.1007/s00773-019-00624-5
B. Ge, H. Zhang, L. Jiang, Z. Li, and M. M. Butt, “Adaptive unscented Kalman filter for target tracking with unknown time-varying noise covariance,” Sensors, vol. 19, no. 6, Art. no. 1371, Mar. 2019, doi: 10.3390/s19061371. [Online]. Available: https://doi.org/10.3390/s19061371
J. Zheng, D. Yan, M. Yan, Y. Li, and Y. Zhao, “An unscented Kalman filter online identification approach for a nonlinear ship motion model using a self-navigation test,” Machines, vol. 10, no. 5, Art. no. 312, May 2022, doi: 10.3390/machines10050312. [Online]. Available: https://doi.org/10.3390/machines10050312
D. Selimovic, J. Lerga, J. Prpic-Oršic, and S. Kenji, “Improving the performance of dynamic ship positioning systems: A review of filtering and estimation techniques,” J. Mar. Sci. Eng., vol. 8, no. 4, Art. no. 234, Apr. 2020, doi: 10.3390/JMSE8040234. [Online]. Available: https://doi.org/10.3390/JMSE8040234
E. Chauhan, M. Sirswal, D. Gupta, and A. Khanna, “A critical review: SANET and other variants of ad hoc networks,” in Proc. Int. Conf. Innov. Comput. Commun. (ICICC 2020), vol. 2, pp. 1093–1114, 2020.

S. Hamad, Y. H. Ali, and S. H. Shaker, “Localization technique model of ships ad hoc network (SANET) using geographic’s database and clustering analysis,” Int. J. Online Biomed. Eng., vol. 18, no. 6, pp. 47–64, 2022, doi: 10.3991/ijoe.v18i06.29819. [Online]. Available: https://doi.org/10.3991/ijoe.v18i06.29819
S. Hamad, K. M. Ali Alheeti, Y. H. Ali, and S. H. Shaker, “Clustering and analysis of dynamic ad hoc network nodes movement based on FCM algorithm,” Int. J. Online Biomed. Eng., vol. 16, no. 12, pp. 47–69, 2020, doi: 10.3991/ijoe.v16i12.16067. [Online]. Available: https://doi.org/10.3991/ijoe.v16i12.16067
G. Abdulameer and Y. H. Ali, “Enhanced clustering techniques for marine ad-hoc networks using dynamic PFCM,” Fusion Pract. Appl., vol. 18, no. 1, pp. 261–268, 2025, doi: 10.54216/FPA.180118. [Online]. Available: https://doi.org/10.54216/FPA.180118
X. Wang, Y. Zhang, Z. Liu, S. Wang, and Y. Zou, “Design of multi-modal ship mobile ad hoc network under the guidance of an autonomous ship,” J. Mar. Sci. Eng., vol. 11, no. 5, Art. no. 962, May 2023, doi: 10.3390/jmse11050962. [Online]. Available: https://doi.org/10.3390/jmse11050962
L. Zhao, G. Shi, and J. Yang, “Ship trajectories pre-processing based on AIS data,” J. Navig., vol. 71, no. 5, pp. 1210–1230, Sep. 2018, doi: 10.1017/S0373463318000188. [Online]. Available: https://doi.org/10.1017/S0373463318000188
T. Lv, P. Tang, and J. Zhang, “A real-time AIS data cleaning and indicator analysis algorithm based on stream computing,” Sci. Program., vol. 2023, 2023, doi: 10.1155/2023/8345603. [Online]. Available: https://doi.org/10.1155/2023/8345603
X. Wang, X. Song, Z. Yan, and Y. Cheng, “Target positioning based on AIS data and optical remote sensing image fusion,” SPIE, May 2024, Art. no. 10, doi: 10.1117/12.3029102. [Online]. Available: https://doi.org/10.1117/12.3029102
M. H. Assaf, G. Voicu, and E. M. Petriu, “The use of Kalman filter techniques for ship track estimation,” WSEAS Trans. Syst., vol. 19, pp. 7–13, Feb. 2020, doi: 10.37394/23202.2020.19.2. [Online]. Available: https://doi.org/10.37394/23202.2020.19.2
R. Skulstad, G. Li, T. I. Fossen, B. Vik, and H. Zhang, “Dead reckoning of dynamically positioned ships: Using an efficient recurrent neural network,” IEEE Robot. Autom. Mag., 2019, doi: 10.1109/MRA.2019.2918125. [Online]. Available: https://doi.org/10.1109/MRA.2019.2918125
B. Cole, M. R. Benjamin, and S. Randeni, “AIS-based collision avoidance in MOOS-IvP using a geodetic unscented Kalman filter,” in Proc. OCEANS Conf., 2021, doi: 10.23919/OCEANS44145.2021.9705900. [Online]. Available: https://doi.org/10.23919/OCEANS44145.2021.9705900
E. A. Wan and R. Van Der Merwe, “The unscented Kalman filter for nonlinear estimation,” in Proc. IEEE 2000 Adapt. Syst. Signal Process., Commun., Control Symp. (AS-SPCC), pp. 153–158, 2000. doi: 10.1109/ASSPCC.2000.882463. [Online]. Available: https://doi.org/10.1109/ASSPCC.2000.882463
B. Cole and G. Schamberg, “Unscented Kalman filter for long-distance vessel tracking in geodetic coordinates,” arXiv:2111.13254, Nov. 2021, doi: 10.48550/arXiv.2111.13254. [Online]. Available: https://doi.org/10.48550/arXiv.2111.13254
Y. Wang, L. P. Perera, and B. M. Batalden, “Particle filter based ship state and parameter estimation for vessel maneuvers,” in Proc. Int. Offshore Polar Eng. Conf., pp. 2722–2728, 2021.
Y. Lu, K. Yang, D. Yang, H. Ding, J. Weng, and R. W. Liu, “Graph learning-driven multi-vessel association: Fusing multimodal data for maritime intelligence,” arXiv:2504.09197, Apr. 2025, doi: 10.48550/arXiv.2504.09197. [Online]. Available: https://doi.org/10.48550/arXiv.2504.09197
Y. Xiao, X. Luo, T. Wang, and Z. Zhang, “Spatio-temporal transformer networks for inland ship trajectory prediction with practical deficient automatic identification system data,” Appl. Sci., vol. 14, no. 22, Art. no. 10494, Nov. 2024, doi: 10.3390/app142210494. [Online]. Available: https://doi.org/10.3390/app142210494
Z. Fan, D. Shen, Y. Bao, K. Pham, E. Blasch, and G. Chen, “RNN-UKF: Enhancing hyperparameter auto-tuning in unscented Kalman filters through recurrent neural networks,” in Proc. 27th Int. Conf. Inf. Fusion (FUSION), 2024, doi: 10.23919/FUSION59988.2024.10706523. [Online]. Available: https://doi.org/10.23919/FUSION59988.2024.10706523
Y. Miao, B. Li, and X. Guo, “Research on multi-submersible positioning prediction based on Kalman filter and LSTM neural network,” in J. Phys.: Conf. Ser., 2024, doi: 10.1088/1742-6596/2898/1/012026. [Online]. Available: https://doi.org/10.1088/1742-6596/2898/1/012026

Details

Primary Language

English

Subjects

Artificial Intelligence (Other)

Journal Section

Research Article

Authors

Anfal Ibraheem
0009-0008-1442-0889
Iraq

Sumaya Hamad ^*
0000-0002-0849-9526
Iraq

Publication Date

September 30, 2026

Submission Date

January 27, 2026

Acceptance Date

June 22, 2026

Published in Issue

Year 2026 Volume: 9 Number: 4

DOI

https://doi.org/10.35377/saucis...1873223

IZ

https://izlik.org/JA62ZD54MW

Cite

RIS / Bibtex

APA

Ibraheem, A., & Hamad, S. (2026). A Hybrid UKF–ST-Graph Transformer Framework for Cooperative Localization in Ship Ad-Hoc Networks. Sakarya University Journal of Computer and Information Sciences, 9(4), 1220-1230. https://doi.org/10.35377/saucis...1873223

AMA

1.Ibraheem A, Hamad S. A Hybrid UKF–ST-Graph Transformer Framework for Cooperative Localization in Ship Ad-Hoc Networks. SAUCIS. 2026;9(4):1220-1230. doi:10.35377/saucis.1873223

Chicago

Ibraheem, Anfal, and Sumaya Hamad. 2026. “A Hybrid UKF–ST-Graph Transformer Framework for Cooperative Localization in Ship Ad-Hoc Networks”. Sakarya University Journal of Computer and Information Sciences 9 (4): 1220-30. https://doi.org/10.35377/saucis. 1873223.

EndNote

Ibraheem A, Hamad S (September 1, 2026) A Hybrid UKF–ST-Graph Transformer Framework for Cooperative Localization in Ship Ad-Hoc Networks. Sakarya University Journal of Computer and Information Sciences 9 4 1220–1230.

IEEE

[1]A. Ibraheem and S. Hamad, “A Hybrid UKF–ST-Graph Transformer Framework for Cooperative Localization in Ship Ad-Hoc Networks”, SAUCIS, vol. 9, no. 4, pp. 1220–1230, Sept. 2026, doi: 10.35377/saucis...1873223.

ISNAD

Ibraheem, Anfal - Hamad, Sumaya. “A Hybrid UKF–ST-Graph Transformer Framework for Cooperative Localization in Ship Ad-Hoc Networks”. Sakarya University Journal of Computer and Information Sciences 9/4 (September 1, 2026): 1220-1230. https://doi.org/10.35377/saucis. 1873223.

JAMA

1.Ibraheem A, Hamad S. A Hybrid UKF–ST-Graph Transformer Framework for Cooperative Localization in Ship Ad-Hoc Networks. SAUCIS. 2026;9:1220–1230.

MLA

Ibraheem, Anfal, and Sumaya Hamad. “A Hybrid UKF–ST-Graph Transformer Framework for Cooperative Localization in Ship Ad-Hoc Networks”. Sakarya University Journal of Computer and Information Sciences, vol. 9, no. 4, Sept. 2026, pp. 1220-3, doi:10.35377/saucis. 1873223.

Vancouver

1.Anfal Ibraheem, Sumaya Hamad. A Hybrid UKF–ST-Graph Transformer Framework for Cooperative Localization in Ship Ad-Hoc Networks. SAUCIS. 2026 Sep. 1;9(4):1220-3. doi:10.35377/saucis. 1873223