摘要
为有效评价交通信号控制的效果,提出了一种基于特征图谱的评价方法.首先,采用两阶段K-means聚类对车辆运行速度轨迹进行聚类,得到描述速度轨迹特征的交通模式;然后,根据模式吸引度、路段评价指数、分布均衡度等指标构建交通信号控制评价的特征图谱,直观体现交通流演变的时空规律.最后,以北京市道路为例进行验证.结果表明,基于特征图谱的评价方法能够细致、直观地描绘交通流的演变过程和运行状态,进而对信号控制进行有效地评价,具有良好的实用性.
In order to effectively evaluate the effect of traffic signal control,an evaluation method based on feature maps was proposed. Firstly,two-stage K-means clustering is used to cluster the vehicle speed trajectories,and a traffic pattern describing the characteristics of the speed trajectory is obtained;then,the traffic signal control evaluation is constructed according to the indexes such as model attraction,road evaluation index,and distribution equilibrium. Feature maps intuitively reflect the temporal and spatial patterns of traffic flow evolution. Finally,this research performed an empirical analysis of roads in Beijing. The results show that the evaluation method based on feature map can describe the evolution process and operation state of traffic flow in detail and intuitively,and then effectively evaluate the signal control,which has good practicability.
作者
张琳琳
李华明
卓为
张耀伟
郑淑欣
李振龙
ZHANG Linlin;LI Huaming;ZHUO Wei;ZHANG Yaowei;ZHENG Shuxin;LI Zhenlong(Beijing University of Technology,Beijing 1000124,China;Beijing Municipal Commission of Transport,Beijing 100073,China;Beijing Traffic Managment Bureau,Beijing 100037,China)
出处
《交通工程》
2019年第6期8-12,共5页
Journal of Transportation Engineering
基金
北京市交通行业科技项目(北京市交通委员会)资助
关键词
交通控制
信号控制
信号控制评价
交通模式
特征图谱
traffic control
signal control
signal control evaluation
trafficpattern
feature atlas
