首页|针对自适应巡航的驾驶风格识别方法研究

针对自适应巡航的驾驶风格识别方法研究

扫码查看
为提高自适应巡航系统对不同风格驾驶员的适用性,将驾驶员的驾驶风格融入自适应巡航控制算法中,通过实车进行驾驶数据采集并分析,获得可表征驾驶风格的特征参数,利用K-Means算法对所有驾驶员的特征参数进行聚类,采用BP神经网络进行驾驶风格识别,提出了一种结合K-Means算法与BP神经网络的驾驶风格识别方法.仿真实验表明,该方法可以将不同驾驶员识别为谨慎、一般、激进三种风格,且识别准确率在 90%以上.研究结果可为自适应巡航控制系统提供更加精准的控制策略,提高系统的适用性和可靠性,使不同风格的驾驶员都能够获得更好的驾驶体验.
Research on Driving Style Recognition Method for Adaptive Cruise Control
In order to improve the applicability of the adaptive cruise system to different styles of drivers,the driver's driving style was integrated into the adaptive cruise control algorithm,the driving data was collected and analyzed by the real vehicle to obtain the characteristic parameters that could characterize the driving style,the K-Means algorithm was used to cluster the characteristic parameters of all drivers,and the BP neural network was used for driving style recognition,and a driving style recognition method combining K-Means algorithm with BP neural network was proposed.Simulation experiments show that the proposed method can identify different drivers as cautious,average,and aggressive styles,and the recognition accuracy is more than 90%.The research results can provide more precise control strategies for adaptive cruise control,improve the applicability and reliability of the system,and enable drivers of different styles to obtain a better driving experience.

driving stylecluster analysisBP neural network

杨新朋、李凡、张丽萍

展开 >

辽宁工业大学 汽车与交通工程学院,辽宁 锦州 121001

驾驶风格 聚类分析 BP神经网络

2024

辽宁工业大学学报(自然科学版)
辽宁工业大学

辽宁工业大学学报(自然科学版)

影响因子:0.226
ISSN:1674-3261
年,卷(期):2024.44(6)