Martin Arndt
ETSI's Martin Arndt discusses recent and current ITS standards development
As the calendar rolls over, and we hop from continent to continent and World Congress to World Congress, where Memoranda of Understanding and cooperation agreements are the headline news, it is easy for those not intimately involved to forget that standards definition is a well-nigh continual process. Significant progress has been made in recent months towards achieving the critical mass and economies of scale which are going to drive development and deployment in, amongst other things, cooperative infrastructures and Martin Arndt, Technical Officer with responsibility for ITS within the European Telecommunications Standards Institute (ETSI), details an extensive list of standards which have been published or are being progressed towards formalisation.Modular architecture
"Perhaps one of the most important pertaining to intelligent transport systems is a European standard for a communications architecture, EN 302 665, which features a modular structure. This allows the various ITS implementations which have different communications requirements to share harmonised interfaces. EN 302 665 was published in September this year; with that"In practical terms, for instance, it doesn't matter if an ITS service uses WLAN or UMTS to communicate because its content is harmonised and reaches the other layers in a standardised form. Pushing back from the infrastructure side, it also means, for example that information from a local or city council providing information on parking availability can be shared quite readily with other ITS systems."
5.9GHz developments
In January, ETSI also published a harmonised standard for the use of the 5.9GHz band."This provides the technical regulation which will avoid harmful interference and lead to more efficient use of the frequency band," Arndt explains. "This standard, along with the previously announced frequency allocation, forms the legal basis for putting 5.9GHz-based systems on the market in Europe. While the general arrangements aren't exactly the same as those in the US and Japan, the differences aren't great enough to preclude deployment of the same systems in each market.
"The EU and the American and Japanese governments have all stressed the need for harmonisation. The ball is certainly rolling - Australia, for example, has recently announced that it will allocate the 5.9GHz frequency band to ITS and there's the potential for other countries to follow."
Greater cooperation
Cooperative infrastructures have also made major strides forward in some key areas.
June 2009 saw publication of a Taking a view on HMIs
"It's going to be very interesting to see how ICT technologies develop in relation to the vehicle. We've seen various HMI solutions emerge in recent years, including LCD-based displays and windscreen projection. However, if we get away from concentrating on the hardware and look more to the software then we can start to use the dash in a much more flexible way because we don't have a static environment any more. That'll come with the cooperative environment: drivers can say, 'I don't want speed, I want a map.' OEMs, meanwhile, can gain differentiation - luxury vehicles can have more features, for example. I don't think that that's too unreasonable to expect."catalogue of a basic set of 90 use cases as defined by cooperative infrastructure stakeholders - applications such as intersection and collision warnings, or emergency vehicle approach. Work has been done to synchronise that set and build what are termed Cooperative Awareness Messages (CAMs).
"We've also worked to define a decentralised environmental notification service; such geo-networking standards are important for how you or more correctly your vehicle is supplied with information relevant to your local environment. Technically, it's possible to share data between vehicles in two different cities - Paris and Brussels, for instance. But would you want to do that? Would the information be at all relevant? More locally, in a dense urban setting, an emergency vehicle might cross your path or a breakdown might occur in close proximity to you but in such a way that neither would necessitate you to take action.
"The requirements have been published and there is an architecture. This lays out the basics for how things work and the timing requirements, especially those for safety-critical applications.
Now, we're looking at how IP packages can be transmitted; with the basic specifications published we now need to look at how IPv6 and other protocols can be integrated. I'd expect to see results from that by mid-2011." Other, spectrum-related activity has looked at the power control mechanisms in 5.9GHz-based technologies with a view to reducing interference, promoting efficient use and avoiding spectrum congestion.
"In an urban environment, a transmission range of 200-300m is usually sufficient from the infrastructure side. Other information, such as that on braking, would be transmitted by the vehicles themselves to others in the vicinity. Information on a broken-down vehicle might be passed between it and other vehicles in close proximity. You pass by, your vehicle picks up the information and passes it on to others, which in turn do the same. But only to those affected. This chain of point-to-point links, of hopping messages, means that only a few cars speak to each other, not many thousands, and a clear characteristic of mesh networks is that you don't need high-powered transmission equipment. I'd expect standards for this to be finalised by the end of 2010." Security is a huge concern going forward. With very large numbers of senders and receivers of data in dense environments it is crucial to ensure that hacking or spoofing does not occur. March 2010 saw another major milestone reached, in the form of an ETSI report which gave an analysis of threat vulnerabilities as well as which security technologies, such as encryption, need to be implemented.
That works two ways, according to Arndt: "You need to be sure that the car or piece of infrastructure that you receive information from is what and where it claims to be. Also, there's a lot of information floating around inside the vehicle, such as that on recent journeys and movements, that for privacy reasons you might not want people to know." Another major piece of the puzzle, in terms of the standards themselves, is the need to get in place a testing base standard in order to ensure that the automotive OEMs all have a common base to work from such that (for instance)
Placing faith
What we'll end up with, says Arndt, is a situation where trusted systems can be implemented in vehicles which (in line with the Vienna Convention) greatly assist but do not take over the driving function."In terms of the standards we need to have an environment populated with information, we're almost there now," he continues. "Now it's just a case of populating that environment with vehicles. The automotive OEMs now need to look at providing a mechanism for updating onboard applications such that people with older vehicles don't become 'lost' in that information environment. That means finding a way of getting ICT [Information and Communication Technology] lifecycles into vehicles' lifecycles.
"Currently there's no true standardisation effort going on for nomadic devices. The current strategy is to get a basic standard of information into the vehicle onboard systems, where there's already a lot of data amassed. OEMs are reluctant to provide an interface to nomadic devices that would allow them to access the sorts of data held on a vehicle's databus.
"Nomadic devices are maturing for applications such as navigation but the HMI [Human-Machine Interface - see Sidebar, 'Taking a view on HMIs'] issues aren't yet sorted. Even if you could achieve very deep integration of the nomadic device with the dashboard, you still have identification issues. OEMs will include a GSM, UMTS or LTS module in a vehicle but how do you CAM onto a nomadic device and so authorise it onto a network? The use of SIM card data might be a solution but I'm not yet convinced that the nomadic device is ready to be an additional processor in the cooperative environment; providing legacy vehicles with a harmonised interface for cooperative applications remains a challenge.
"The current objective is to fit all new cars with cooperative systems but carmakers will only include such systems if their cost point is such that they can effectively be written off and 'lost' in the vehicle. That's why we need harmonisation: to achieve the necessary economies of scale."
Wider issues
In terms of the overall health of the standardisation effort, Arndt feels that the historical difficulty of getting the various automotive OEMs around the same table has been addressed."We've got that understanding we needed: that common technology is an advantage to all. We now have a strong, useful standardisation effort in the sense that with a large number of companies involved we have technological commonality.
"The issue now is how to address the policy framework issues: how to get standards in place - spectrum allocation and so on - that are also supported by local administrations. We need to increase awareness of the fact that everyone benefits. If a driver can for example drive all the way from the UK via France to Morocco using the same service, then we have something that even politicians at the local level can sell. Safety doesn't recognise geographical borders.
"Market factors also have an important role to play - GSM rollout in the US, which facilitated global roaming, came about as a result of consumer pressure. On the flipside, the manufacturing side, we have to recognise that the automotive industry is truly global, with the major OEMs turning out products on several continents. They want to use the same systems and components globally." There is however an outstanding discussion: that over whether standards are always necessary. HMIs are a case in point, says Arndt.
"Vehicles don't need to have the exact same HMIs and there is still space here for marque differentiation and branding. On the infrastructure side, we've had two decades of ITS development which has led to many island solutions. Cooperative infrastructures aren't in place yet so we have an opportunity to streamline things globally. If that's possible, then we should definitely go for it."
