报告摘要: In intelligent transportation systems (ITS), traffic state estimation is a fundamental work for traffic prediction, traffic control, dynamic route guidance, traffic incident detection, traffic safety surveillance, etc. However, until now, it is still very difficult to accurately estimate traffic states for large-scale urban road networks in real time. The reason mainly lies in the following aspects: 1) In contrast with freeways or urban arteries, urban road networks bear many complex factors to impact traffic flows, e.g. traffic signals, different turning rates at intersections, and vehicles appearing or disappearing on roads; 2) The stop-and-go behavior of vehicles brings great difficulties for an accurate measurement and estimation of the traffic flows; 3) It is a big challenge in both theory and engineering practice to continuously estimate the traffic flow variations for large-scale urban road networks, which usually contain tens of thousands of road sections.

In this report, several traffic state estimation methods will be introduced, supported with results of real-data experiments. The estimation methods use single-source traffic information from different detectors, such as loop detectors, GPS probe vehicles, or multi-source traffic information from multiple detectors. For the method with GPS probe vehicles, two different algorithms will be further discussed and be compared with each other using plenty of ground truths of real links. Finally, several promising directions will be proposed for the future research.

报告人简介: Qing-Jie Kong (M’07) received the PhD degree in pattern recognition and intelligent systems from Shanghai Jiao Tong University, Shanghai, China, in 2010. From 2008 to 2009, he was a visiting scholar with the Beckman Institute for Advanced Science and Technology, the Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL. From 2010 to 2012, he was a Postdoctoral Scientist with the Department of Automation, Shanghai Jiao Tong University. Since 2012, he has been an Assistant Professor with the State Key Laboratory for Management and Control of Complex Systems, Institute of Automation, Chinese Academy of Sciences, Beijing, China. He has authored more than 40 publications in international journals, book chapters, and conference proceedings, and applied more than 10 patents. He is an associate editor for the IEEE Transactions on Intelligent Transportation Systems. His research interests include traffic data mining and fusion, traffic network modeling and analysis, and video object detection and recognition.