高速路相邻前车驾驶人意图预测  

Prediction for Driver’s Intention in Adjacent Front Vehicles on Expressway

作　　者：袁小慧[1,2] 归文强 YUAN Xiaohui;GUI Wenqiang(School of Automobile,Chang’an University,Xi’an 710064,China;School of Vehicle Engineering,Xi’an Aeronautical Institute,Xi’an 710077,China)

机构地区：[1]长安大学汽车学院,陕西西安710064 [2]西安航空学院车辆工程学院,陕西西安710077

出　　处：《测试技术学报》2023年第2期93-98,共6页Journal of Test and Measurement Technology

基　　金：国家自然科学基金资助项目(51775053);西安航空学院校级科研资助项目(2020KY1230)。

摘　　要：针对相邻车道前车并道至自车与本道前车之间会使自车的跟车间距突然减小,进行高速公路实车试验,采集自车、自车所在车道以及相邻车道车辆的运动状态参数,基于SVM算法和Fisher判别法对高速公路相邻前车的驾驶人意图进行预测。结果表明,自车车速、自车与当前车道前车的车头时距、纵向相对速度以及自车与相邻前车的纵向相对速度可以作为相邻前车驾驶意图的表征参数,所建立的驾驶意图预测模型能够提前2 s预测驾驶人的意图,且SVM模型和Fisher模型的驾驶意图预测准确率分别为94.1%和91.1%,该模型能够为自适应跟车控制提供一定的理论指导。Aiming at the sudden reduction of the following distance because of the lane change of the vehicle in the adjacent lane,the real vehicle test on the expressway is carried out.The motion state parameters of the vehicles in the lane where the vehicle is located and the vehicles in the adjacent lane are collected,and the driver's intention of the adjacent vehicle in front is predicted based on SVM algorithm and Fisher discriminant method.The results show that four parameters can be used as the characterization parameters of the driving intention of the adjacent front vehicle,including the ego-vehicle speed,the headway and the longitudinal relative speed between the ego-vehicle and the vehicle in front of the current lane,the longitudinal relative speed between the ego-vehicle and the adjacent front vehicle.The established driving intention prediction model can predict the driver's merging intention 2 s in advance.The driving intention prediction accuracies for SVM model and Fisher model are 94.1%and 91.1%,respectively.The model can provide some theoretical guidance for adaptive car following control.

关 键 词：交通安全 意图预测 SVM FISHER 并道行为 

分 类 号：U491.253[交通运输工程—交通运输规划与管理]