Deep learning technologies for predicting ship routes

This article examines the mechanisms for using deep learning to predict ship routes. As maritime traffic volumes rapidly increase, effective forecasting is a critical task for intelligent transportation systems. Deep learning (DL) is a subset of machine learning that uses multilayer artificial neural networks to automatically detect complex patterns in big data. Unlike classical machine learning, DL models independently extract features (feature learning) without human intervention, enabling high accuracy in computer vision, speech recognition, and generative AI. Deep learning enables computational models consisting of multiple processing layers to learn data representations with multiple levels of abstraction.

1. Li, M. W., Han, D.F., Wang, W.L. Vessel traffic flow forecasting by RSVR with chaotic cloud simulated annealing genetic algorithm and KPCA // Neurocomputing. – 2015. – P. 243–255.

2. He, Y. et al. An improved Kalman filter model for short-term vessel traffic flow prediction // Journal of Navigation. – 2017.

3. Wang, Ch., Zhang, X., Chen, X., Li, R., Li, G. Vessel traffic flow forecasting based on BP neural network and residual analysis // 4th International Conference on Information, Cybernetics and Computational Social Systems (ICCSS). – 2017. – Pр. 350–354.

4. Yang, L., Hao, Y., Liu, Q., Zhu X. Ship traffic volume forecast in bridge area based on enhanced hybrid radial basis function neural networks // Proceedings of the International Conference on Transportation Information and Safety (ICTIS), Wuhan, China, 25–28. – June 2015. – Pр. 38–43.

5. Chen, L. et al. Stacked LSTM (long short-term memory) network for traffic congestion forecasting on realistic navigation maps // IEEE Transactions on Intelligent Transportation Systems. – 2016.

6. Lv, Y., Duan, Y., Kang, W., Li, Z., Wang, F. Y. Traffic flow prediction with big data: A deep learning approach // IEEE Transactions on Intelligent Transportation Systems. – 2015. – 16(2). – Pр. 865–873.

7. Zhang, J., Zheng, Y., Qi D. Deep spatio-temporal residual networks for citywide crowd flow prediction // Proceedings of the AAAI Conference on Artificial Intelligence (AAAI-17). – 2017.

8. Zhao, L., & Liu, Y. (2023). Hybrid Deep Learning Models for Real-time Ship Collision Avoidance and Path Planning. Reliability Engineering & System Safety, 231. – Рр. 108–124.

9. Li, S. et al. A Transformer-based Framework for Vessel Trajectory Prediction with Attention Mechanisms // Ocean Engineering. – 2022. – 244. – Pр. 110–135.

10. Wang, X. et al. Spatiotemporal Graph Convolutional Networks for Vessel Traffic Flow Forecasting in Complex Waterways // IEEE Transactions on Intelligent Transportation Systems, 2023. – № 24(8). – Pр. 8567–8582.

11. Zhang, M. et al. Diffusion Models for AIS Data Recovery and Trajectory Generation in Marine Navigation // Journal of Marine Science and Engineering. 2024. – № 12(2). – P. 341.