摘要
车辆换道行为是机动车在道路行驶时的一种常见行为,对其进行预测能有效减少交通事故。为准确预测目标车辆换道行为,根据车辆在时间内的横向位移,提出基于多阈值的车辆换道预测方法。该方法根据法规,结合车速设置换道时的横向位移阈值、方向盘转角与转速阈值,探索多阈值间的相互作用关系,通过多阈值协同预测驾驶员换道意图。结果表明,系统的预判准确率可达95%以上,说明该方法能准确预测车辆换道意图,且实时性较好,能为辅助驾驶技术的发展提供一定参考。
Vehicle lane changing behavior is a common behavior of motor vehicles while driving on the road,and predicting it can effectively reduce traffic accidents.To accurately predict the lane changing behavior of the target vehicle,a multi threshold based vehicle lane changing prediction method is proposed based on the lateral displacement of the vehicle over time.This method is based on regulations and combines the lateral displacement threshold,steering wheel angle,and speed threshold when changing lanes at vehicle speed.It explores the interaction relationship between multiple thresholds and predicts the driver's intention to change lanes through collaborative prediction of multiple thresholds.The results show that the prediction accuracy of the system can reach over 95%,indicating that the method can accurately predict the vehicle's intention to change lanes and has good real-time performance,which can provide a certain reference for the development of assisted driving technology.
作者
冯雅
孟凡盛
刘洋洋
FENG Ya;MENG Fansheng;LIU Yangyang(College of Information Engineering,Shaoxing Vocational Technical College,Shaoxing,Zhejiang 312000,China;College of Information Engineering,Huaibei Institute of Technology,Huaibei,Anhui 235000,China;School of Engineering,Anhui Agricultural University,Hefei,Anhui 230036,China)
出处
《自动化应用》
2024年第1期32-34,共3页
Automation Application
基金
浙江省访问工程师项目(FG2022194)。
关键词
多级阈值
车辆工程
换道预测
multi-level threshold
vehicle engineering
lane changing prediction
