EUVIMEDEUVIMED entdecken
Crossref

A Deep Learning Framework for Vessel Destination Prediction in Maritime IoT Systems

Jin Chen, Ryan Wen Liu, Wu Zhou, Maohan Liang, Shengzheng Wang, Chang Peng

IEEE Internet of Things Journal · 2026-11 · 13 (22) · 51915-51930

DOI: 10.1109/jiot.2026.3728540 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Internet of Things Journal, 2026-11. Verfasst von Jin Chen und 5 weiteren Autor:innen. Die Arbeit nennt 53 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Hainan Provincial Natural Science Foundation of China, 2024 Open Research Fund of the Key Laboratory of Transport Industry of Marine Technology and Control Engineering.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The transition toward maritime decarbonization and intelligent shipping has increased the demand for Maritime Internet of Things (IoT) systems capable of supporting traffic coordination, smart-port operations, and route-planning assistance. However, inaccurate or missing destination information in automatic identification system (AIS) data remains a persistent obstacle to reliable maritime decision-making. To address this problem, this study proposes a deep learning framework that reformulates vessel destination prediction as a spatial image-classification task. The framework converts observed AIS trajectories into fixed-view images within a common geographic coordinate system, while excluding the reported AIS destination field and class-specific port markers from the model input. A multiscale residual convolutional neural network is then employed to capture both local route details and broader spatial patterns. An auxiliary trajectory-image reconstruction task is further introduced to regularize the learned representation and preserve informative route structures. Experiments conducted on a real-world AIS dataset from Danish waters demonstrate that the proposed framework achieves superior predictive performance compared with the evaluated baseline methods. Additional robustness and ablation analyses confirm the effectiveness of its principal components. These findings demonstrate the potential of spatial trajectory-image learning for vessel destination prediction and its applicability to Maritime IoT and smart-port operations.

Förderung

  • National Natural Science Foundation of China
  • Hainan Provincial Natural Science Foundation of China
  • 2024 Open Research Fund of the Key Laboratory of Transport Industry of Marine Technology and Control Engineering

Zitieren

APA

Chen, J., Liu, R. W., Zhou, W., Liang, M., Wang, S., & Peng, C. (2026). A Deep Learning Framework for Vessel Destination Prediction in Maritime IoT Systems. IEEE Internet of Things Journal, 13(22), 51915-51930. https://doi.org/10.1109/jiot.2026.3728540

Vancouver

Chen J, Liu RW, Zhou W, Liang M, Wang S, Peng C. A Deep Learning Framework for Vessel Destination Prediction in Maritime IoT Systems. IEEE Internet of Things Journal. 2026;13(22):51915-51930. doi:10.1109/jiot.2026.3728540

BibTeX
@article{Chen2026,
  author = {Chen, Jin and Liu, Ryan Wen and Zhou, Wu and Liang, Maohan and Wang, Shengzheng and Peng, Chang},
  title = {A Deep Learning Framework for Vessel Destination Prediction in Maritime IoT Systems},
  journal = {IEEE Internet of Things Journal},
  year = {2026},
  volume = {13},
  number = {22},
  pages = {51915--51930},
  doi = {10.1109/jiot.2026.3728540}
}
Kostenlos starten →