With the development of the global economy and the increase of international trade, the role of the shipping industry has become more and more important. However, the complex maritime transportation environment brings great challenges to ship safety management, and frequent collisions make it especially critical to accurately predict ship trajectories. Currently, multi-model fusion techniques have been widely used to improve prediction accuracy, and the Transformer model has begun to be used in ship trajectory prediction. Although Transformer performs well in dealing with the global dependence of long sequences, it does not have its own structure specialized in dealing with time series and may face information loss when dealing with very long sequences. In this study, we enhance the temporal characteristics of the input data by introducing sLSTM as a preprocessing step to provide Transformer with richer inputs with temporal characteristics and to improve its processing ability for time-sensitive tasks. Experiments are conducted using real historical AIS data, and the results show that the method significantly outperforms mainstream methods in predicting ship trajectories in complex maritime environments, with improved accuracy and prediction performance.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    A hybrid sLSTM-transformer network for ship trajectory prediction


    Contributors:

    Conference:

    Third International Conference on Geographic Information and Remote Sensing Technology (GIRST 2024) ; 2024 ; Rome, Italy


    Published in:

    Proc. SPIE ; 13551


    Publication date :

    2025-04-02





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Ship trajectory prediction based on transformer model

    Kang, Kaihang / Zhang, Chuang / Guo, Chen | IEEE | 2022


    Ship trajectory prediction model based on transformer

    Li, Na / Huang, Yaoliang / Jiang, Dapeng | SPIE | 2023


    TrAISformer: Spatio-Temporal Ship Trajectory Prediction Based on Transformer

    Li, Yunbo / Wang, Jiayu / Li, Tao et al. | IEEE | 2024