原文传递
A Method for Detecting Incidents from Traffic Detector Data based on the Non-parametric Statistics
| 题名： | A Method for Detecting Incidents from Traffic Detector Data based on the Non-parametric Statistics |
| 其他题名： | 1)船岡直樹,佐藤光,割田博,桑原雅夫:首都高速道 路における突発的ボトルネック判定手法構築に関する研究,交通工学研究発表会論文集,Vol.29,pp.301-304,2009. |
| 正文语种： | 日文 |
| 作者： | 成岡 尚哉 |
| 关键词： | Incident detection;Non-parametric model;Traffic detector |
| 摘要： | Fast and precise detection of traffic incidents is required to mitigate negative effects by using traffic management measures. This study proposes an incident detection method that detects an incident from observation data of traffic detectors by using non-parametric statistics derived from historical observation data. One of the characteristics of the non-parametric statistics is that the method can work without any trial-and-error parameter calibrations. The proposed method detects rare occurrence events that seem to be caused by traffic incidents, and consists of the two components namely estimating occurrence probability of traffic state based on the non-parametric statistics and checking consistency with a traffic state caused by an incident bottleneck. The proposed model was empirically verified by using actual field data obtained from of the proposed method was comparable to the California algorithm and that the proposed model had high applicability due to simplicity of parameter calibration. |
| 出版年： | 2015 |
| 论文唯一标识： | C-226Y2015V50N01022 |
| 中文栏目名称： | 交通工学論文文集 第1卷 第1号 |
| 英文栏目名称： | JSTE Journal of Traffic Engineering Vol. 1,No.1 |
| 论文中文关键词： | 突発的ボトルネック検出;ノンパラメトリックモデル;車両感知器 |
| 期刊名称： | 交通工学 |
| 拼音刊名(出版物代码)： | C-226 |
| 卷： | 50 |
| 期： | 01 |
| 页码： | 后插11-后插20 |
相关文献
- Automatic Incident Detection for Urban Expressways Based on Segment Traffic Flow Density
- PERFORMANCE EVALUATION OF STATE TRANSPORT UNDERTAKINGS USING NON-PARAMETRIC TECHNIQUES
- Estimating Fundamental Diagram for Signalized Intersections Using Connected Vehicle Data
- Comparison of different Bayesian methods for estimating error bars with incident duration prediction
- Large-Scale Loop Detector Troubleshooting Using Clustering and Association Rule Mining
- Visualizing Traffic Dynamics Based on Floating Car Data
检索历史
应用推荐
