中文摘要：

目前,国内城市出行信息服务平台和网络地图平台主要提供静态信息服务,城市交通动态特征难以得到真实反映,对以时间和费用为主要准则的公众出行缺乏实用价值。本文提出了历史数据推理和微观交通仿真相结合,进行短时交通预测以服务公众出行的一种新方法;设计了实时交通信息处理与发布服务器、GIS应用服务器与数据库管理系统三者协同工作的体系结构;实现了顾及短时预测交通状况的公众出行路径规划过程,并作了验证。因此,为公众动态出行信息服务和动态网络电子地图的实现,提供了一个可行的解决方案。

英文摘要：

Implementing convenient traveling information service is a crucial task for carrying out the strategy of intelligent transportation system and developing location-based services.At present,most of the domestic traveling information service systems only provide relatively static information which can＇t reflect the possible short-term changes of traffic,and result in very limited practical use.Although there have emerged some car navigation products and other applications involving real-time traffic information,considering the rapid change of city traffic situation,these products and applications still face practical difficulties for all the information received real-timely will get outdated within a few minutes,which makes the so called dynamic applications basically time-slice limited static ones.This paper presents a practical short-term traffic prediction approach in real-time conditions by integrating historical traffic based statistical reasoning with back propagation neural network based analytical model,and commercial microscopic traffic simulation software.The historical traffic based statistical reasoning utilizes the inherent traffic rules by identifying the general spatial-temporal distribution pattern,and ignores any inputs that don＇t contribute to the output during the training process,including the gross error in the collected traffic data.The commercial microscopic traffic simulation software process the traffic abnormities that always exist in big cities,and has no requirement for the collected real-time traffic information to cover the whole road network,hence provides an effective supplement for short-term traffic forecast.Then an approach is developed combining GIS server,traffic prediction server and database management system to implement dynamic route guidance.The traffic prediction server receives real-time traffic information obtained from floating vehicles and achieves short-term forecasting results for the whole road networks,then fed the results back into the database management syst