The invention relates to a CNN (Convolutional Neural Network), improved LSTM (Long Short Term Memory) and attention mechanism-based ship trajectory prediction method and system, computer equipment and a storage medium. The method comprises the following steps: extracting ship trajectory data through data processing; a forgetting gate and an input gate of the LSTM model are fused, the cell state at the last moment is reserved to serve as an F-LSTM model, a convolutional neural network and an attention mechanism are introduced, and a preliminary ship trajectory prediction model is generated; and performing parameter adjustment on the initial ship trajectory prediction model according to the ship trajectory data, and performing ship trajectory prediction through the target ship trajectory prediction model. By extracting the ship trajectory data, convenience is provided for subsequent model training; by combining the CNN, the improved LSTM and the model trained by the attention mechanism, the spatial features and the time sequence features of the trajectory data can be fully mined, different weights are allocated to different features through the attention mechanism, useless feature information is filtered, and the precision of ship trajectory prediction is improved.

    本发明涉及一种基于CNN、改进LSTM及注意力机制的船舶轨迹预测方法、系统、计算机设备及存储介质。所述方法包括:通过数据处理提取船舶轨迹数据;融合LSTM模型的遗忘门和输入门,并保留上一时刻的细胞状态,作为F‑LSTM模型,并引入卷积神经网络以及注意力机制,生成初步船舶轨迹预测模型;根据船舶轨迹数据对初步船舶轨迹预测模型进行参数调整,通过目标船舶轨迹预测模型进行船舶轨迹预测。通过提取出船舶轨迹数据,为后续模型训练提供了便捷;结合CNN、改进LSTM及注意力机制训练完成的模型,可以充分挖掘轨迹数据的空间特征和时序特征,并且通过注意力机制为不同的特征分配不同的权重,过滤无用的特征信息,提高船舶轨迹预测的精度。


    Access

    Download


    Export, share and cite



    Title :

    CNN, improved LSTM and attention mechanism-based ship trajectory prediction method and system


    Additional title:

    基于CNN、改进LSTM及注意力机制的船舶轨迹预测方法、系统


    Contributors:
    LI XIULAI (author) / LIU BOYI (author) / CHEN MINGRUI (author) / ZHANG BIN (author) / PENG XIN (author) / LU WENHAI (author)

    Publication date :

    2024-04-12


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    Chinese


    Classification :

    IPC:    G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS / G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    A Bi-directional LSTM Ship Trajectory Prediction Method based on Attention Mechanism

    Zhang, Sheng / Wang, Long / Zhu, Mingdong et al. | IEEE | 2021



    Ship Trajectory Prediction with Social History LSTM

    Zhao, Wenfeng / Zhang, Xudong | IEEE | 2023


    Probabilistic Vehicle Trajectory Prediction Based on LSTM Encoder-Decoder and Attention Mechanism

    Meng, Dejian / Zhang, Lijun / Xiao, Wei et al. | SAE Technical Papers | 2022


    RF and LSTM combined ship trajectory prediction model

    ZHANG CONG / ZHU JISHUAI / DENG MEIHUAN et al. | European Patent Office | 2024

    Free access