Through their traffic managers, authorities are having to demonstrate tangible evidence that their traffic-reduction measures are having a positive effect. One way of achieving this is through the use of a sophisticated Journey Time Measurement System (JTMS) using Automatic Number Plate Recognition (ANPR) from Siemens.
Traditional Urban Traffic Control systems have always been able to provide an indication of congestion and journey times, but the use of Automatic Number Plate Recognition (ANPR) technology provides a better, more accurate mechanism for monitoring journey times to provide a meaningful measure of overall network performance.
Minimising the disruption that occurs when an incident takes place is central to providing an efficient, well-managed network.
The integration of journey time information into a Comet UTMC
system can provide incident detection capabilities to the operator in the control room. By comparing the current ‘live’ journey time with a historical profile, the system can automatically alert operators to abnormal conditions, allowing prompt action to be taken. At the same time, Siemens’ Comet JTMS provides quality information, to both other authorities and highway users. This enables drivers to avoid the problem area rather than waste time sitting in a jam.
JTMS is proving to be a multi-faceted solution, enabling traffic managers to report back on long-term performance and justify spending to councillors and the public. It also provides a platform for the broadening horizons of driver information provision, integrated traffic management solutions and crime fighting.
Nigel Weldon from Siemens explains the merits of using a JTMS; “A Journey Time Measurement System allows traffic managers to build up a picture of trends over time. Having the numbers to back up valuable resources spent on congestion reducing initiatives is something that’s becoming increasingly important for local authorities, and that’s exactly what JTMS can offer.”
The traffic management bill places many demands upon local authorities for the efficient management and control of their highway network. Automatic Number Plate Recognition technology provides a new and accurate mechanism for monitoring journey times and presents the most accurate and meaningful measure of overall network performance.
The data collected can also be used in the calculation of on-line origin and destination matrices to provide the local authority with a complete, up-to-date model of the major traffic flows in the urban environment for modelling new traffic management schemes. In addition, the data collected by the camera network is invaluable to the user in providing a continual review of traffic flows on all major routes into and out of an urban area.
It is no longer necessary to carry out expensive, manpower-intensive surveys at infrequent intervals to determine the major traffic slows within the network for accurate modelling. The integration of ANPR journey time information into a UTMC common database such as Comet provides a complete performance overview for efficient network management
The primary challenge of journey time monitoring with ANPR is to obtain a high match rate of vehicles passing between the camera pairs, while minimising the physical number and optimising the location of cameras.
On inter-urban routes, obtaining a good match rate is relatively easy, and origin destination patterns can be predicted with reasonable confidence. Within the urban environment, this is much less easy to achieve - drivers have many more options for exiting the main roads, making it especially important where you locate the ANPR cameras to obtain a high enough match rate.
Installations in Essex on dual-carriageway routes have typically looked at Lane 1 only, although in some locations where there is a well defined split of the traffic, such as at an approach to a roundabout, two lanes are monitored to ensure a high capture rate of traffic continuing both straight on and turning.
As part of the project, floating vehicle data was compared with ANPR data for journey times on a specific link on the A127. The floating vehicle data consistently calculated a faster journey time than the ANPR-based calculation. This led to questions about the accuracy of the ANPR data. However, the ANPR data consistently gave a journey time which equated to an average speed of around 55mph.
As this road is a dual carriageway and typically Lane 1 is being monitored, the reason was immediately obvious - a high match rate was being obtained but this was primarily from heavy goods vehicles. For long-term monitoring, this is not an issue as the overall trends will still be visible, and for incident detection, vehicles in Lane 1 are just as likely to be affected as vehicles in any other lane.
If there is not enough traffic passing a particular detection point, regardless of whether it is a perfect site with power and access readily available, then the match rate will be poor and the data unreliable. This is much more noticeable in the urban environment, where the location of the ANPR camera is absolutely critical to ensure an acceptable match rate for journey time calculation.
Another issue which is critical to achieving a good match rate is the location and field of view of the camera. In some cases in Essex, a difference in location of 20 yards has been a determining factor in whether or not a suitable match rate has been achieved, as a result of traffic behaviour on the approach to or just after the chosen location.
Other issues which must be borne in mind include subtle failures which can affect the accuracy of the system, such as camera alignment issues, and also vandalism of cameras (unfortunately, members of the public sometimes confuse them with speed detection cameras). It is important to closely monitor the system to detect those cameras which previously recorded quality data but which are now reporting a suspiciously low match rate and check them out.
Siemens’ Comet system assists the operators in the detection of subtle failures by monitoring the quality of the data and highlighting automatically when large variances occur. Overall, the use of ANPR to provide journey time information in Essex has been a valuable addition to the overall traffic management strategy, although not without issues which needed to be addressed as the project has progressed.
As with many systems, the optimum use is only gained when a solid understanding of the capability, as well as any limitations is known. This has been particularly important with the JTM system when locating camera sites and ensuring that the traffic flows and origin-destination patterns are appropriate to provide a high match rate for accurate journey time calculation.
2010-Sep-24 |