The invention discloses a multi-stage and multi-mode vehicle trajectory prediction method, which comprises the following steps: S1, collecting vehicle data, and constructing a data set based on a data sample; s2, dividing a data set into a training set, a verification set and a test set; s3, establishing a first model to predict an initial end point, correcting the initial end point based on the lane line, generating a candidate end point based on the corrected initial end point, and establishing a second model based on the candidate end point to generate a multi-modal trajectory; s4, training the first model and the second model; and S5, inputting the data in the test set into the first model to generate an initial end point, generating a plurality of candidate end points based on the initial end point and the lane line, generating a plurality of candidate tracks based on the candidate end points and the second model, and outputting a predicted track based on the candidate tracks and a top-k algorithm. The method has the beneficial effects that the predicted trajectory of the vehicle can be restrained in a reasonable range by means of the lane line, so that the predicted trajectory is accurate and safe.

    本发明公开了一种多阶段多模态的车辆轨迹预测方法,包括以下步骤:S1:采集车辆数据,基于数据样本构建数据集;S2:数据集划分为训练集、验证集和测试集;S3:建立第一模型预测初始终点,基于车道线修正初始终点,基于修正后的初始终点生成候选终点,基于候选终点建立第二模型生成多模态轨迹;S4:训练第一模型和第二模型;S5:测试集中的数据输入到第一模型生成初始终点,基于初始终点和车道线生成若干个候选终点,基于候选终点和第二模型生成若干条候选轨迹,基于候选轨迹和top‑k算法输出预测轨迹。本发明的有益效果是:能借助车道线将车辆预测的轨迹约束在合理的范围内,使预测出的轨迹准确且安全。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Multi-stage and multi-mode vehicle track prediction method


    Weitere Titelangaben:

    一种多阶段多模态的车辆轨迹预测方法


    Beteiligte:
    CHEN QIHUANG (Autor:in) / XIAO ZHONGWEN (Autor:in) / ZHANG ZHEN (Autor:in) / LI BOXUAN (Autor:in) / WEN SICHAO (Autor:in) / LI KAIBIN (Autor:in)

    Erscheinungsdatum :

    2022-09-02


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    G06Q Datenverarbeitungssysteme oder -verfahren, besonders angepasst an verwaltungstechnische, geschäftliche, finanzielle oder betriebswirtschaftliche Zwecke, sowie an geschäftsbezogene Überwachungs- oder Voraussagezwecke , DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES / G06K Erkennen von Daten , RECOGNITION OF DATA / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G08G Anlagen zur Steuerung, Regelung oder Überwachung des Verkehrs , TRAFFIC CONTROL SYSTEMS



    Multi-Stage Linear Prediction Model for Railway Track Irregularity

    Guo, Ran / Han, Bao-Ming | Tema Archiv | 2013


    Target vehicle track prediction method for multi-lane scene

    ZUO ZHIQIANG / GUO SONGLIN / WANG YIJING et al. | Europäisches Patentamt | 2023

    Freier Zugriff

    MULTI-STAGE SWITCHABLE INERTIA TRACK ASSEMBLY

    BRADSHAW JEFF / HARTGERS ANDREW | Europäisches Patentamt | 2016

    Freier Zugriff

    Multi-stage switchable inertia track assembly

    Europäisches Patentamt | 2016

    Freier Zugriff

    MULTI-STAGE SWITCHABLE INERTIA TRACK ASSEMBLY

    BRADSHAW JEFF / HARTGERS ANDREW | Europäisches Patentamt | 2018

    Freier Zugriff