Eric Sampson, visiting professor at Newcastle University and City University London and ambassador for
Twenty-some years is more than long enough for the complexion of an industry to change completely. And so it is with ITS; the directions in which it has moved and the delivery channels for systems and services are distinctly different from what was envisaged when the concept of applying technology to improve road networks' operation and management was first being given serious consideration. Which is not to suggest that all is not well with the world; although there remain some significant obstacles to progress to be overcome, the fact is that a great deal of progress has been, and is continuing to be, made.
Perhaps the most significant change, says Eric Sampson, visiting professor at Newcastle University and City University London and ambassador (and past chairman) of ITS-UK, is that technology has now passed out of the laboratory and into the high street.
"Or, to put it another way, we've seen a bunch of 'stuff', disparate elements which only existed as objects of interest and curiosity at World Congresses, coalesce into sellable systems and services. We now have more than passable satnavs, for example, whose prices have fallen whilst their performance and battery lives have increased substantially," he says.
"Some problems have got worse - certainly congestion has increased. The recent recession might have given pause to its growth but it'll be back because everyone wants a car. And both local and national government aren't prepared yet to grasp the need for Road User Charging [RUC] although they're going to have to in the not-too-distant future.
"I'd say that the timeframes for delivery and convergence that were being talked about 20 years ago were way out but now, although there has been no real, concerted effort towards its creation, an open architecture is pretty much there.
"Deployment channels have changed radically with the intrusion of consumer electronics and I see an interesting fight developing between three groups, each of which is looking to monopolise the market. The automotive manufacturers take the view that, because the wireless interface resides in the vehicle, the service delivery market belongs to them. The PDA and smartphone people will have it that the automotive manufacturers 'only' supply a 'socket', a physical interface through which their products can interact with the consumer, so the market is theirs. The WiFi and telecoms people in turn feel that because location and bonding with the vehicle happens through wireless then they have ownership.
"The word 'consumer' doesn't appear anywhere in this and we need someone to take all three, bang their heads together and force some form of agreement - although who that might be isn't clear. The OECD is a possibility.
"So we'd assumed two delivery routes, the car companies and the road operators, when we in fact have three with the arrival of the phone companies, who are in reality providers of apps. I think it's a very healthy development but we need some things to change. Government departments hold huge amounts of data relevant to transport and it's time to allow someone with legitimate access to get in there and mash and correlate that into something more useful. Around the world, we need DOTs and other government bodies to improve access to data. I know that the UK DfT believes this and is trying to deliver but not everyone is supporting them."
Gains and losses
Moore's Law pretty much guarantees the impossibility of accurately predicting systems' capabilities a decade or so out, if only because entrepreneurial spirits will intervene and create whole new industries where there was neither the space nor means of delivery before. Advancement and achievement in the ITS space is undeniable - as are the gaps. Not all of the latter are in the hardware and software."Across the board there have been major technological gains: think about the progress made with GPS, telecommunications, the services which are delivered to personal devices, active traffic management and the evolution of smart journey planners," Sampson continues.
"There are 'holes', though, principally in the soft issues. Behavioural issues have profound effects on deployment and we've barely scratched the surface here. Driver distraction is acknowledged to be a problem but driver underload has barely been considered as yet. As we make vehicles smarter, so electronics take on more of the load and, perversely, we have to consider whether we're almost being too safe. In some respects, the holes are being caused by the technology. I'm not convinced that drowsy driver technologies are the whole answer. The automotive companies might say that they'll install a system that'll monitor a driver's blink rate or some-such, and then provide an alert through the audio system or by making the seat vibrate, but that's just accepting the problem and then providing a counter, rather than addressing it in the first instance. Some of the solutions are non-technical - a road with gentle curves in it will provide a driver with more interaction with his or her local environment than a long, straight one. The freight industry in particular is very concerned about this. And, of course, I have to come back to the non-adoption of RUC."
Crying wolf
Like other fledgling industries, ITS has suffered from being both over- and undersold. Speed of deployment, partly through optimism, partly through ignorance, certainly suffered from the former, according to Sampson.
"But we made a mistake. We thought cooperative infrastructure was about robotic driving. It's not: it's about the exchange of information between static and dynamic entities. We assumed the benefit would be the closer stacking of cars and completely ignored the information dimension.
"It's a simple fact that there's very little that today's driver can do to benefit mankind. The only way to have that happen is through large-scale intervention, and we've massively undersold the benefits of wide-scale network management. We've got to do away with in-and-out
lane-changing that disrupts productivity and instead optimise the overall network performance. No-one but the infrastructure operators can do that.
"An obstacle in this respect is politicians' fear of being 'wrong'. If we research something and it fails, that's not a failure as such - we've just proved that something doesn't work. An example is the M42's Active Traffic Management scheme. That faced some very vociferous critics concerned that we'd placed so many 'ugly' VMS gantries so close together along the route. Our answer, of course, was that we didn't know what the optimum spacing should be. It turns out that we can place the VMS gantries farther apart - but if we hadn't tried that, we wouldn't have known."
Architectures
"National ITS architectures need more coordinated effort, especially in Europe, to reduce or remove duplication of effort and move us more into the territory of open systems. It's already happening in smart ticketing, which provides us with a counter to those who say it can't be done."The European Commission's FRAME initiative has resulted in something which has been adopted in France, Italy, the Netherlands and a couple of other places, and I think that needs more impetus. The US has shown what can be done, in many respects. Essentially, it's said, 'If you're spending federal money, you've got to use the national architecture.' On the one hand, it's made manufacturing non-compliant products commercial suicide, on the other it's made sure that plug-and-play is really quite meaningful."
Understanding the gains
There is an assumption that because a lot of people are now talking about cost benefit analyses we have started to realise the importance of selling the output and not the input. As ITS has moved to the deployment stage, it has had to fight for a share of real-world funding. A consequence of this is that various public- and private-sector organisations have been set up to try to ascertain systems' true worth. We are far from having cracked the issue, however; in Sampson's view we are still in the foothills of a very long climb."One drum I'm banging quite hard is over cost-benefit analyses. We assume we've made progress. But go to a typical local authority and look at its pet projects. Quite often, the project with the biggest cost benefit is simply too expensive to implement, so options B, C and D are implemented instead.
"There's no calculus whereby something which actually has a low cost benefit but which enables many other things can be looked at in a wider sense. Issuing smart cards to the population of a town might provide many people with mobility; transport might get a lousy rate of return but in the round the things which happen as a consequence are worth far more. There are examples in the US where some proposed Interstate routes actually had a negative cost benefit but when considered as major diversion routes which improved other parts of the network they then came to have major advantages. Another example might be the installation of a fibre-optic backbone along a strategic network. Taken in isolation, that might have a very poor cost benefit. But when you consider the things you can hang off it - traffic monitoring and management systems, traveller information and so on - things look quite different.
"We need to think in terms of a system of systems because if you take something out in some place there'll be compensations elsewhere. This is an area which needs a lot of hard work. We can't consider things piecemeal - we have to start looking at sets of initiatives.
"Perhaps also we need a new vocabulary. We're still selling technical performance, not benefit to the traveller. We should be selling guaranteed journey times, quality, reliability and resilience.
"Liability continues to bedevil us and I think the motor industry is being dishonest. It's using the old Vienna Convention, which I feel is obsolete, and which states that a driver must have control of his or her vehicle at all times , to argue that we meet this requirement with an 'on-off' switch for applications such as intelligent cruise control or lane following. Liability is a bogey and the car companies need a stern talking to and to be brought into line. Understandably, there are concerns over liability and being sued but the triplicate fly-by-wire systems used by aircraft since the 1970s prove that liability and safety need not be an issue. These work by majority vote; if all three systems are saying the same thing, then no problem. If it becomes a two-against-one issue, then the pilot is alerted. Cost is sometimes put up as a barrier. Fine; but 'low cost' is not the same as 'low price', and the components used, for example, in vehicles' electronic stability control systems are essentially consumer electronics components - they don't cost vast amounts. Again, I think we'll see trucks as a starting point for progress."
The coming years
The goalposts have moved since we started down the path of ITS, Sampson notes. "That was in the days before modern IT, GSM, the likes of Google, social networking sites such as Facebook and Twitter. The future is going to be about much greater personalisation of services and I think that's embodied in the 'Internet of things'. At present, we access the 'Net through a PC, a laptop or a smartphone. But we'll see a point where all kinds of devices have connectivity. So what does that mean? Well, it means that your car can take account of all the traffic data it receives, ascertain a realistic journey time - which can change in real time - and tell your house that you're almost home. On a cold winter's day, the garage might already be open, the lights on and the oven warming, all without you having to do anything."If everyone shared their intentions, traffic network management would be a lot easier. If an individual was to tell the network that at a given time he or she will be undertaking a journey between Point A and Point B, the network can take account of this. The personalised information pushing back to the car might include a message like, 'The traffic where you are is slowing but don't worry, your journey is unaffected.' A more comprehensive database of incidents would also be useful, so that the network knows the consequences of a given type of incident at a given location. The level of intervention can then be ratcheted up or down accordingly.
"The message to the road user in all of this is that we don't need to know who you are; just tell us what you want to do and when. Governments need to get over their neuroses concerning privacy.
"RUC seems to be to be inevitable. Politicians are simply going to have to realise that with congestion up and fuel and other revenues down, there's a different and better way of doing things.
"And when does all this happen? I think a decade is too soon for e-Safety applications such as vehicle platooning. It's not too soon to expect to see other forms of vehicle-infrastructure integration, however."
