The invention discloses a train full-speed measurement method based on a deep learning multi-source information fusion algorithm, and the method comprises the steps: obtaining a train operation speed data set at a prediction moment, inputting the train operation speed data set into a pre-trained train speed measurement model, and outputting a train operation speed result; the train speed measurement model is obtained by training based on multi-sensor fusion detection train speed data; the training process of the train speed measurement model comprises the steps that a train speed data set collected by multiple sensors in a plurality of time periods is obtained and divided into a training set and a verification set, the training set is used for training the train speed measurement model, and the verification set is used for train speed measurement model optimization. According to the invention, by adopting a multi-source information fusion algorithm based on deep learning, the characteristics of three kinds of sensor data of the binocular camera, the magnetic steel and the millimeter-wave radar are fused, so that the running speed of the train is predicted, high-precision speed measurement of the train in a full speed range of 0-400km/h is realized, and the speed measurement precision of the train is improved. The speed range covers the highest speed of the train in the current rail transit field.

    本发明公开了基于深度学习多源信息融合算法的列车全速度测量方法,包括:获取预测时刻的列车运行速度数据集,将其输入预先训练好的列车测速模型,输出列车运行速度结果;所述列车测速模型是基于多传感器融合检测车速数据训练得到;所述列车测速模型的训练过程包括:获取若干时段多传感器采集的车速数据集,并将其划分为训练集和验证集,所述训练集用于训练列车测速模型,所述验证集用于列车测速模型调优。本发明通过采用基于深度学习多源信息融合算法,融合了双目相机、磁钢和毫米波雷达三种传感器数据的特征,从而预测列车的运行速度,实现了对列车在0‑400km/h全速度范围内的高精度测速,该速度范围覆盖了当前轨道交通领域列车的最高速度。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Train full-speed measurement method based on deep learning multi-source information fusion algorithm


    Weitere Titelangaben:

    基于深度学习多源信息融合算法的列车全速度测量方法


    Beteiligte:
    WEI ZHIPENG (Autor:in) / TIAN HAILONG (Autor:in) / CHEN SHILIN (Autor:in) / HE XUYANG (Autor:in) / LU XIN (Autor:in)

    Erscheinungsdatum :

    2024-04-02


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06F ELECTRIC DIGITAL DATA PROCESSING , Elektrische digitale Datenverarbeitung / B61L Leiten des Eisenbahnverkehrs , GUIDING RAILWAY TRAFFIC / G01P MEASURING LINEAR OR ANGULAR SPEED, ACCELERATION, DECELERATION OR SHOCK , Messen der Linear- oder Winkelgeschwindigkeit, der Beschleunigung, der Verzögerung oder des Stoßes / G01S RADIO DIRECTION-FINDING , Funkpeilung / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen



    Metro train speed measurement method based on multi-source information intelligent fusion

    LU HAITING / FU BAOMING / CHEN LINSHAN et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Multi-sensor fusion maglev train ground speed measurement method and system

    LIU BIN / ZOU JINJIAN / GONG WENZHONG et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    Train accurate positioning system based on multi-source information fusion technology

    LIU MEI / FANG MENGZHI / LUO WEI et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    RAIL TRAIN FULL-REDUNDANCY SPEED MEASUREMENT METHOD AND SYSTEM

    LIANG XINGXING / QIN YEJUN | Europäisches Patentamt | 2023

    Freier Zugriff

    Rail train full-redundancy speed measurement method and system

    LIANG XINGXING / QIN YEJUN | Europäisches Patentamt | 2023

    Freier Zugriff