Until recently, measured attributes of vehicles or road traffic have depended on available equipment and technology, rather than necessarily being the most useful indicators. This situation is now changing. For example, accurate vehicle data are now available, which for the first time enable travel time to be measured for each vehicle. Now that typical travel times in free flow can be known, any significant time increase can be measured, and its difference can be described as delay time. Thus accurate travel time measurement can assist incident detection and management, and significantly reduce frustrating and harmful delays. To measure travel times over a road network directly, measurement sites need to be set up over the network, and it is necessary to recognise individual vehicles consistently at each site. New data acquisition techniques, developed by consultants WS Atkins and Diamond Consulting Services as part of their Idris incident detection work, enable any vehicle to be identified with error probability 0.00005 in free-flowing traffic and 0.0002 in congested traffic, at a realistic cost. They enhance the output of a standard vehicle detector and generate an accurate high-quality parametric description of each vehicle. They offer significant benefits for all traffic management systems.
Abstract