交通运输系统工程与信息 ›› 2013, Vol. 13 ›› Issue (5): 120-126.

所属专题: 车路协同与智能化技术

• 决策论坛 • 上一篇    下一篇

基于变精度粗糙集的汽车碰撞危险态势评估

彭理群1,2,吴超仲*1,2,黄珍3   

  1. 1.武汉理工大学 智能运输系统研究中心,武汉 430063; 2.水路公路交通安全控制与装备教育部工程研究中心,武汉430063;3.武汉理工大学 自动化学院,武汉 430070
  • 收稿日期:2013-04-01 修回日期:2013-05-13 出版日期:2013-10-25 发布日期:2013-11-08
  • 作者简介:彭理群(1984-),男,湖北武汉人,博士生.
  • 基金资助:

    国家自然科学基金(61104158,51178364);教育部新世纪优秀人才计划(NCET-10-0663);武汉市学科带头人计划(201271130445).

Situation Assessment of Vehicle Collision Risk Based on Variable Precision Rough Set

PENG Li-qun1,2, WU Chao-zhong1,2, HUANG Zhen3   

  1. 1.Intelligent Transport Systems Research Center, Wuhan University of Technology, Wuhan 430063, China;2.Engineering Research Center of Transportation Safety, Ministry of Education, Wuhan 430063, China;3. Automation school, Wuhan University of Technology,Wuhan 430070, China
  • Received:2013-04-01 Revised:2013-05-13 Online:2013-10-25 Published:2013-11-08

摘要:

汽车碰撞危险辨识与预警是智能防撞系统的关键技术之一,为了解决现有的防撞系统在复杂交通环境下虚警率较高、灵活性差的问题,本文对“人—车—路”多因素影响下的汽车碰撞危险辨识方法进行了研究.综合考虑驾驶员、车间距、路面等因素对行车安全性的影响,并基于车路协同平台获取相关信息,应用态势评估理论建立汽车碰撞危险评估模型.在该模型的基础上,结合变精度粗糙集理论形成汽车碰撞危险态势评估规则.应用属性加权相似度方法比较当前行车状态与决策信息表中所有行车状态的相似程度,得到碰撞危险态势的评估结果.模拟驾驶实验结果表明,该方法能融合行车安全相关的多种因素来检测碰撞风险,为汽车防撞系统提供准确的决策.

关键词: 智能交通, 态势评估, 变精度粗糙集, 汽车防撞系统, 人-车-路多因素, 车路协同

Abstract:

Vehicle collision risk identification and warning is one of the key technologies of intelligent collision avoidance system. In view of the problems that the existing vehicle collision avoidance systems keep high false alarm rate and low flexibility in complicated road traffic environment, this paper presents a method for vehicle collision risk identification with the impact from “drivervehicleroad” multifactors. A model is established for identification of vehicle collision risk considering the fusion of related factors such as driver state, distance between vehicles, road surface, etc. The relevant information is obtained from cooperative vehicleinfrastructure system (CVIS). Then, the risk situation assessment algorithm is formulated based on theory of variable precision rough set (VPRS). Finally, similarity degrees between the current driving status and driving status in decisionmaking table are compared based on attribute weighted similarity, which could get the situation assessment results. The simulated driving results show that this method can be used for fusion of safety related factors and detection of collision risk.

Key words: intelligent transportation, situation assessment, variable precision rough set (VPRS), vehicle collision avoidance system, drivervehicleroad multifactors, cooperative vehicleinfrastructure system (CVIS)

中图分类号: