The invention provides a multi-style driver forced lane change prediction method fusing evolutionary game and machine learning, and the method comprises the steps: extracting a feature vector representing the game state of a lane change vehicle and surrounding vehicles and a vehicle lane change decision observation value based on vehicle trajectory data, and building a forced lane change observation data set; dividing vehicles into two driving styles by using a Gaussian hybrid clustering method, dividing lane changing games into four classes according to different driving styles of lane changing vehicles and vehicles behind a target lane, and dividing an observation data set into four corresponding subsets; according to the vehicle trajectory data of the four subsets, outputting a lane changing decision by using an evolutionary game theory (EGT) and a collision time index; according to the method, a forced lane change model EGT-ML fusing an evolutionary game theory and machine learning is established, an EGT-based lane change decision is added into a training set, model training is performed on the training set, a prediction effect is evaluated on an observation data set, and a result shows that the model has relatively high accuracy and relatively strong robustness.

    本发明提供一种融合演化博弈与机器学习的多风格驾驶者强制换道预测方法,该方法包括基于车辆轨迹数据提取表示换道车辆与周围车辆博弈状态的特征向量及车辆换道决策观测值,建立强制换道观测数据集;利用高斯混合聚类方法,将车辆分为两种驾驶风格,根据换道车辆和目标车道后车的不同驾驶风格,将换道博弈分为四类,将观测数据集划分为相对应的四个子集;根据四个子集的车辆轨迹数据,分别利用演化博弈论EGT及碰撞时间指标输出换道决策;建立融合演化博弈论和机器学习的强制换道模型EGT‑ML,将基于EGT的换道决策加入训练集中,并在训练集上进行模型训练,在观测数据集上评价预测效果,结果表明模型具有较高的准确率和较强的鲁棒性。


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    Multi-style driver forced lane change prediction method fusing evolutionary game and machine learning


    Weitere Titelangaben:

    一种融合演化博弈与机器学习的多风格驾驶者强制换道预测方法


    Beteiligte:
    WANG CHEN (Autor:in) / LI MENGYUN (Autor:in) / XU SIXUAN (Autor:in) / ZHOU WEI (Autor:in) / SHI MINGLIANG (Autor:in)

    Erscheinungsdatum :

    2023-06-23


    Medientyp :

    Patent


    Format :

    Elektronische Ressource


    Sprache :

    Chinesisch


    Klassifikation :

    IPC:    B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion



    Lane changing device and method for autonomously learning lane changing style of driver

    LAI FENG / KANG DAIKE / WANG SIQUAN et al. | Europäisches Patentamt | 2021

    Freier Zugriff

    Driver Lane Change Prediction Using Physiological Measures

    Kochhar, Dev S. / Watta, Paul / Wang, Tianyu et al. | SAE Technical Papers | 2015


    Driver lane change prediction using physiological measures

    Murphey,Y.L. / Kochhar,D.S. / Watta,P. et al. | Kraftfahrwesen | 2015


    Driver Lane Change Prediction Using Physiological Measures

    Murphey, Yi lu / Kochhar, Dev S. / Watta, Paul et al. | British Library Conference Proceedings | 2015


    Driver Monitoring-Based Lane-Change Prediction: A Personalized Federated Learning Framework

    Du, Runjia / Han, Kyungtae / Gupta, Rohit et al. | IEEE | 2023