This study presents a framework for highway incident detection using vehicle kinetics, such as speed profile and lane changing behaviour, as envisioned in the vehicle-infrastructure integration (VII, also known as IntelliDrive) in which vehicles and infrastructure communicate to improve mobility and safety. This framework uses an in-vehicle intelligent module, based on a support vector machine (SVM), to determine the vehicle's travel experiences with autonomously generated kinetics data. Roadside infrastructure agents (also known as RSUs: roadside units) detect the incident by compiling travel experiences from several vehicles and comparing the aggregated results with the pre-selected threshold values. The authors developed this VII-SVM incident detection system on a previously calibrated and validated simulation network in rural Spartanburg, South Carolina and deployed it on an urban freeway network in Baltimore, Maryland to evaluate its transportability. The study found no significant differences in the detection performance between the original network and a new network that the VII-SVM system has not seen before. This demonstrated the feasibility of developing a generic VII-SVM system, applicable across transportation networks.
Accelerated incident detection across transportation networks using vehicle kinetics and support vector machine in cooperation with infrastructure agents
IET Intelligent Transport Systems ; 4 , 4 ; 328-337
2010-12-01
10 pages
Article (Journal)
Electronic Resource
English
British Library Conference Proceedings | 2010
|Traffic Incident Detection Using Multiple-Kernel Support Vector Machine
Transportation Research Record | 2012
|Incident detection using support vector machines
Online Contents | 2003
|Traffic Incident Detection Using Multiple-Kernel Support Vector Machine
Online Contents | 2013
|