Cooperative infrastructures, cooperative enforcement?

A dozen years from now, will enforcement still be constrained by the legislative thinking which currently prevails? Or will the needs of the wider transport community bring about some welcome changes?

The term 'data fusion' is especially fashionable and particularly relevant when one considers cooperative road infrastructure. When it comes to enforcement and the sensor technologies used, however, it remains more of a novelty.

This is driven in no small part by the inherent conservatism of the enforcement sector. Something of a necessity given the need to preserve the integrity of the evidence chain, it is nonetheless a well-documented source of frustration to systems developers and suppliers and was discussed in the article on the greater utility of digital enforcement systems.

For reasons both altruistic and commercial, systems suppliers argue that the current Type Approval process, at least in many places in Europe, stands in the way of progress when it comes to operating regimes, efficiency and, ultimately, safety. In this issue, Paolo Sodi of 125 Sodi Scientifica makes that very point: that by comparison with the US's 'black box' approach to enforcement, which is very much more results-led, Europe's, which results in every minor change to an existing system having to be re-homologated, is both cumbersome and obstructive.

Divergent opinions

But if Sodi's view highlights a common concern, the detection technologies themselves, and the relative merits of each, provoke as many responses as there are those to whom questions are posed.

To many, the inductive loop is old news. And yet Bob Lees of 529 Diamond Consulting Services, creator and supplier of the 36 Idris system, put forward a very strong case for the loop's continuing employment in a variety of detection, classification and enforcement applications. Most importantly, Lees made the point that the development of loops, and perhaps more importantly the algorithms which derive information from them, will not stand still. Within the enforcement sector, his views have both strong supporters and dissenters. The former argue that the loop's robustness and development potential cannot be overlooked; the latter point to the many supposed advantages of newer video- and laser-based alternatives, not least their non-intrusive nature (that is, the absence of a need to cut into the road surface to install them).

John Harris, European Business Development Director at 112 Redflex Traffic Systems, notes that while the loop continues to have considerable merits the tumbling costs of embedded processing technologies have allowed highly technological solutions to become very cost-effective.

"Radar, laser and video suffer from obstruction and occlusion, whereas loop or lane detection offers precise vehicle capture and identification. But loops aren't mobile. Radar remains the best above-ground non-intrusive technology but application is always going to be a consideration when it comes to technology choice.

Loops are likely to continue as a staple detection technology on a diminishing scale fulfilling specific tasks." Daniel Tijl, Product Manager at 101 Peek Traffic, reinforces that last point.

"For spot enforcement, I consider multiple algorithm measurement by a single detector to be the state of the art. That might be as a result of matching flank signals across a pair of loops or of Doppler calculations from radar technologies. Data fusion is more common for section control, and in-image triggering by OCR cameras is an improving technology in this respect.

"There is a lot of work going on when it comes to radar and video but loop detection still has a lot of unrealised potential. We're currently experimenting with new types of loop configurations and radar. It's important for any manufacturer to maintain flexibility in its portfolio." 113 Redspeed International is another loop devotee. The general view there is that video-based detection suffers from frame rate issues, while radar is costly. The company's new RedFusion multi-point average speed system uses laser technology which can achieve something in the region of an 85 per cent capture rate, according to Marketing Manager Nicki Bradley. "Our R&D team is watching the further improvements to lane-specific lasers very keenly," she adds. "And it's only a matter of time before someone harnesses the power of FPGAs and similar devices for enforcement applications." 1679 Gatso's Director of Sales & Marketing, Edmar van der Weijden, also sees a guaranteed future for loops in the medium term, however others, such as 2185 Robot's Peter Schneider, is more than willing to consign them to the dustbin of history.

Schneider even dismisses the need for data fusion: "3D radar has to be considered the state of the art for speed and red light enforcement," he says. "I just don't see loops having any significant advantage over radar any more. Fused technologies could be an efficient way of offering an independent, secondary speed check, as required in some countries, but in general I don't see a trend emerging." Geoff Collins of average speed (section control) specialist 126 Speed Check Services highlights a trend towards digital video enforcement. This is driving interest in Automatic Number Plate Recognition (ANPR), which is readable by both humans and machines and is not 'blind' like a loop.

"Loops offer a reliable trigger mechanism that won't be fooled by bad weather, vibrations or missing plates. But ANPR still has a lot to offer and I don't see one single technology which can offer a comprehensive solution," he continues.

Future drivers

Digitisation has already given greater impetus to technological development work in the detection area. Another market factor is that rising traffic densities are currently pushing greater numbers of users to specify non-intrusive solutions. John Harris notes that an associated development is the demand for increased functionality: "Different requirements are constantly being evolved by customers but enforcement is a demanding market, requiring precision and reliability. The value of the offering needs to be clearly communicated." That perhaps reflects, finally, a emerging realisation among the wider end-user community of the capabilities of digital systems and the operating paradigms they allow. It all rests on whether the legislatures can be convinced, however. Enforcement and privacy have always been uneasy bedfellows and there are knock-on effects on what is necessary in terms of data acquisition, encryption and storage. This is where the specifics of the detection technology used and the wider capabilities of digital systems start to merge.

"There are certain areas where the fusion of evidential integrity and the additional functionality or information which results is simple to achieve," says Nicki Bradley. "A red light camera could be used for traffic monitoring; a network of speed cameras could be used for intelligence gathering; suppliers would be easily able to provide such hybrid systems were they not hampered by legislation.

"Government agencies need to examine current policies and decide whether these still apply given the encryption and watermarking techniques available to developers. There is already a great deal of traffic enforcement equipment available for adaptation; rather than continuing to bring in new equipment for each application would it not be more advantageous to divert funds to other areas such as educating road users?" Geoff Collins concurs. He feels strongly that the use of enforcement systems for wider traffic management might lead to data fusion becoming more commonplace: "I do think there's a way forward. As long as it is designed and operated appropriately, the evidence chain will be secure."

'Cooperative' enforcement?

Perhaps the real answer requires a 180o view and the influence of factors currently outside enforcement. Future cooperative infrastructures and present-day enforcement share many common obstacles, privacy and data integrity among them. Indeed, Peek, for example, is a partner in the European CVIS and SAFESPOT cooperative research projects. Given that two-way information exchange between vehicles and infrastructure is just around the corner, many of the identification and verification headaches which have so long pained the enforcement and tolling sectors are going to plague the wider transport community in the not-too-distant future. It doesn't require too long a stretch of the imagination to enforcement from within becoming a serious option. The options for data fusion might not lie solely with the infrastructure but could come to include the onboard units and location-specific data which their use implies.