交通运输系统工程与信息

交通运输系统工程与信息 ›› 2016, Vol. 16 ›› Issue (4): 95-100.

• 智能交通系统与技术 • 上一篇    下一篇

面向驾驶辅助系统的换道安全性预测模型研究

倪捷，刘志强*，涂孝军，董非   

  1. 江苏大学汽车与交通工程学院，江苏镇江212013
  • 收稿日期:2016-01-05 修回日期:2016-05-16 出版日期:2016-08-25 发布日期:2016-08-26
  • 作者简介:倪捷（1982-），女，江苏启东人，讲师，博士生.
  • 基金资助:

    国家自然科学基金/ National Natural Science Foundation of China（51108209）；教育部博士点基金/The Ph.D. Programs Foundation of Ministry of Education of China（20113227110014）；江苏省普通高校研究生科研创新计划项目/The Research Innovation Program for College Graduates of Jiangsu Province（CXLX12_0657）.

Safety Prediction Model of Lane Changing Based on Driver Assistance System

NI Jie, LIU Zhi-qiang, TU Xiao-jun, DONG Fei   

  1. School of Automobile and Traffic Engineering, Jiangsu University, Zhenjiang 212013, Jiangsu, China
  • Received:2016-01-05 Revised:2016-05-16 Online:2016-08-25 Published:2016-08-26

PDF

被引次数

11

摘要：

为使换道辅助系统适应动态的道路交通条件，以模拟驾驶试验数据为依据，提 出以目标车道后车的最大减速度和平均减速度为衡量指标，分析换道行为对道路交通安 全的影响程度.通过换道影响的K-means 聚类分析结果将换道数据分为危险、安全、舒适 等3 类，对3 类换道数据中的运行参数进行对比检验，发现自车速度、速度差、车间距、纵 向加速度具有显著差异性；以自车速度、速度差、车间距、纵向加速度为输入特征量，以换 道影响的聚类结果为输出特征量，运用支持向量机（SVM）理论建立换道安全性预测模 型.预测结果表明，对危险换道行为的预测准确率可以达到91.1%，该模型具有较好的预 测准确性.

关键词: 智能交通, 换道安全性预测, SVM, 驾驶辅助系统, 换道影响

Abstract:

In order to adapt the dynamic road traffic conditions for the lane changing assistance system, the maximum and average deceleration of the following vehicle on target lane are presented to analyze the influence degree of lane changing on road traffic safety, based on the simulated driving test data. The data are divided into three categories by K- means, such as danger, safety and comfort. The results show that egovehicle velocity, velocity difference and distance between ego- vehicle and adjacent- following vehicle, the longitudinal acceleration have significant differences by comparing the operating parameters of the three kinds data. Taking ego-vehicle velocity, velocity difference and distance between ego-vehicle and adjacentfollowing vehicle, the longitudinal acceleration as input variables, and the kinds of lane change influence as output variables, the safety prediction model of lane changing is established by SVM. The forecasting result shows that the prediction accuracy of the dangerous lane changing behavior can reach 91.1%, and the model has good prediction effect.

Key words: intelligent transportation, safety prediction of lane changing, SVM, ADAS, lane changing influence

中图分类号: