To alleviate all kinds of difficulties in maritime traffic, relevant agencies have equipped ships with an automatic identification system (AIS) that can transmit data. However, the estimated arrival time in AIS data is generally filled in by the captain according to experience before the ship leaves the port, which is quite different from the actual arrival time. To make scientific use of the existing arrival time data set and effectively reduce the uncertainty of arrival time prediction, In this paper, a combined model of long-term and short-term memory (LSTM) and Kalman filter (KF) is built to predict the arrival time of ships. This model combines the advantages of LSTM in capturing long-term dependence, KF can reduce the influence of noise on the prediction results, 250 ship trips are verified and analyzed, and the experiment proves that LSTM-KF is improved in MAE and MSE compared with LSTM.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Prediction Model of Ship Arrival Time using Neural Network and Kalman Filter


    Beteiligte:
    Zhang, Xinyan (Autor:in) / Zhang, Xining (Autor:in) / Li, Pu (Autor:in)


    Erscheinungsdatum :

    24.02.2023


    Format / Umfang :

    1215996 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Bus Arrival Time Prediction: A Spatial Kalman Filter Approach

    Achar, Avinash / Bharathi, Dhivya / Kumar, Bachu Anil et al. | IEEE | 2020



    Echo State Network Ship Motion Modeling Prediction Based on Kalman Filter

    Peng, Xiuyan / Dong, Huiyuan / Zhang, Biao | British Library Conference Proceedings | 2017