作　　者： ; ; ;

机构地区： 长沙理工大学

出　　处： 《长沙交通学院学报》 2004年第1期58-62,共5页

摘　　要： 高速公路事件自动检测算法是高速公路事件管理系统的核心。目前的事件自动检测算法普遍采用流量、速度和/或占有率作为检测参数,这些参数可由道路主线上的检测器得到。但这种检测方法要求路上埋设有一定密度的检测器。而目前在我国的很多高速公路上都存在着埋设检测器数量少、间距大的现象,以至这些算法在实际应用中检测效果很不理想。针对高速公路检测器的埋设现状,提出了一种新的事件检测算法,将主线检测器得到的信息和收费口处得到的信息进行融合来判断交通事件的发生。仿真试验表明:与传统事件检测算法相比,该算法具有更高的检测率、更低的误报率,可以明显改善检测效果。 Automatic incident detection algorithm is the core of ATM. Now, most of the incident detection algorithms adopt speed, occupancy and/or volume as detection parameter obtained from loop detector. But these algorithms need a number of loop detectors with certain distance. At present, in China there aren't enough loop detectors on the freeways. So the detection effect is limited in practice. As for Chinese freeway is concerned, the authors attempt to develop a new freeway incident detection algorithm based on data fusion. The algorithm combines the traffic information of mainline with the information of tolling station to detect the incident. Finally, as the simulation results shown, the new algorithm for incident detection has a lower false alarm rate and a higher detection rate, and it can improve the detection. The algorithm would have a bright prospect in China.

关 键 词： 事件检测 数据融合 学习矢量量化

领　　域： [交通运输工程] [交通运输工程]