项目名称: 高度城市化地区基于海量出租车轨迹数据面向出行者的路径旅行时间预测
项目编号: No.41271441
项目类型: 面上项目
立项/批准年度: 2013
项目学科: 天文学、地球科学
项目作者: 李响
作者单位: 华东师范大学
项目金额: 75万元
中文摘要: 高度城市化地区交通拥挤现象普遍存在,其给每一位出行者带来的直接问题就是令旅行时间无法掌控,因此,旅行时间预测成为一个具有巨大社会需求的研究课题。传统预测方法多基于来自固定点的交通数据采集设备,其安装和维护的费用高,覆盖面有限,因此,预测对象局限于高速公路、快速路网络或城市道路网络中的特定路段。随着车载定位设备的普及,具更大时空覆盖面的海量车辆轨迹信息逐渐成为交通研究的重要数据源。本课题以此为背景,以一般城市道路网络为研究对象,以海量出租车轨迹数据为主要参考数据,配合各种可能获取并且可能引起预测时段交通拥挤的影响因素,兼顾周期行及偶发性交通拥挤对旅行时间的影响,利用神经网络及贝叶斯预测模型发展长时路径旅行时间预测方法。基于已有的研究和数据积累,课题以上海为例,开展实证研究。其成果可为旅行时间预测增加新的研究思路和手段,为交通出行者提供服务,并进一步丰富了海量车辆轨迹数据的处理方法和应用领域。
中文关键词: 旅行时间预测;时空数据索引;可预测性;;
英文摘要: With respect to the ever-increasing traffic congestion in metropolitan area, travel time for every traveller becomes more and more complicated and irregular. How to predict travel time and make it reliable, therefore, is of great importance to the related researches. Requiring high installation and maintenance costs and having limited service area, most existing approaches to travel time prediction depend on fixed-point-based data collection equipments, which renders the research limited to highway, expressway or specific road segments of urban road network. The wide use of mobile positioning devices makes it possible to obtain a large volume of vehicle trajectory data within a short period, which is growing as an alternative dataset for transportation related researches. Based on the aforementioned works and data collection methods, the present project aims at predicting travel time of long paths on urban road network using large volume trajectory data with neural network and Bayesian prediction model. Additionally, other available relevant factors that cause traffic congestion and the corresponding influences on travel time are considered. In order to validate the proposed approach, an empirical research is then conducted in Shanghai, which could provide a novel method of predicting travel time, and enrich the
英文关键词: Travel time prediction;Spatio-temporal data index;predictability;;