As arterial network incident detection system for use in ATMS/ATIS environment is proposed in this research. The key features of the proposed system lie in its event-based simulation structure and self-learning as well as monitoring functions that enable the traffic control center to minimize the response time, and take essential actions prior to the formation of congestion. Furthermore, it can be integrated with the adaptive traffic signal control and route guidance systems to maximize a network's operational capacity during recurrent and non-recurrent congestion. The proposed system consists of the principal components: a dynamic traffic flow prediction model, an incident identification model, and an incident monitoring process. It is designed not only detect incident, but also severity in real time so that the control center can take proper emergency response actions. In this paper, the framework of the system is presented at first, and then methodologies for developing each component are explained in detail. With the presumed surveillance capabilities in labortory experiments, the proposed system achieves quite effective preliminary results, indicating the promising future of research along this direction.
A self-learning system for real-time incident detection and severity assessment - framework and methodology
Ein selbstlernendes System zur Zwischenfallerkennung und Schadenbewertung in Echtzeit-Rahmen und Methodik
1993
8 Seiten, 3 Bilder, 12 Quellen
Conference paper
English
Automotive engineering | 1993
|British Library Conference Proceedings | 1993
|Real Time Freeway Incident Detection
NTIS | 2014
|Continuous Learning Framework for Freeway Incident Detection
British Library Conference Proceedings | 1998
|Continuous Learning Framework for Freeway Incident Detection
Online Contents | 1998
|