Development of Traffic Diagnosis Model to Support Traffic Control System applying Artificial Intelligence Model

应用人工智能模型开发交通诊断模型以支持交通控制系统

• 批准号: 23760494
• 负责人: NISHIUCHI Hiroaki
• 金额: $ 2.66万
• 依托单位: Nihon University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Young Scientists (B)
• 财政年份: 2011
• 资助国家: 日本
• 起止时间: 2011 至 2013
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-23760494/
• 关键词: 交通工学; 交通管制; ベイジアンネットワーク; 車両感知器データ; 交通事故; 国際研究者交流(香港(予定))

This research investigates to apply Bayesian Network technique to develop Traffic Diagnosis Model which detect traffic congestion caused by incidents on Tokyo Metropolitan Expressway. Especially this research focused to detect traffic congestion caused by traffic accident using traffic detector data with learning past data to estimate conditional probability distributions of occurrence the situation by traffic accident at Yoyogi entrance section on route No. 4 of Tokyo Metropolitan Expressway. The hit ratio of detection of traffic congestion by traffic accident was around 40% which is almost same level of accuracy compared with existing research, but the analysis shows the requirements to improve accuracy of proposed model compared with existing study. And it is also recommended to consider traffic condition of whole of expressway network.

本研究旨在应用贝叶斯网络技术开发交通诊断模型，对东京都高速公路突发事件造成的交通拥堵进行检测。特别是，本研究重点利用交通检测器数据结合学习过去的数据来检测交通事故造成的交通拥堵，以估计东京都快速路4号线代代木入口段交通事故发生的条件概率分布。交通事故检测交通拥堵的命中率在40%左右，与已有研究的准确率基本持平，但分析表明，与已有研究相比，本文提出的模型精度还有待提高。并建议考虑整个快速路网络的交通状况。

项目成果

都市内高速道路における交通現象診断モデル構築のための基礎的検討

城市快速路交通现象诊断模型构建基础研究

• 发表时间: 2013
• 作者: Kobayashi;K.;Surjono;Ari;I.R.D;真田純子;西内裕晶,割田博
• 通讯作者: 西内裕晶,割田博