摘要
为弥补基于传统交通信息采集技术的高速公路交通事件检测算法仅能判断交通事件发生与否而无法判断车辆个体是否受到交通事件影响的不足,作者提出了一种基于RFID和FOA-GRNN的高速公路交通事件对车辆影响的判断模型。该模型使用RFID获取关键交通信息,利用广义回归神经网络对交通信息进行归类,应用果蝇优化算法获取广义回归神经网络的最佳平滑参数值。用VISSIM进行了仿真,对模型进行了验证,结果表明该判断模型具有检测率高、误警率低的特点,能迅速判断出受到交通事件影响的车辆,为交通疏导工作提供支持。
In order to compensate the disadvantage of AID based on traditional traffic statistics technology can detect the incident but can't identify whether the individual vehicle is affected by the incident,this article presents a judge model of impact of lane closure incident on individual vehicles on freeways based on RFID technology and FOA-GRNN method.This model uses GRNN to classify the traffic information collected by RFID,uses FOA to optimize the spread value of GRNN.With VISSIM simulation,results shows that the model has the feature of high DR,low FAR,and can quickly detect the vehicle influenced by incident to support traffic evanesce.
出处
《武汉理工大学学报》
CAS
CSCD
北大核心
2012年第3期63-68,共6页
Journal of Wuhan University of Technology
基金
国家自然科学基金(51078232/E0807)
上海交通大学海洋工程国家重点实验室项目(GKZD010027)