Technology convergence, non-linear thinking help V2X architectures come of age
April 15, 2015
With the groundwork being laid for the vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications infrastructure of the future, Scott...
With the groundwork being laid for the vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications infrastructure of the future, Scott McCormick, President of the Connected Vehicle Trade Association (CVTA) and advisor to U.S. Secretary of Transportation Anthony Foxx gives a brief history of the connected vehicle industry, helps define the V2X communications landscape, and describes how the move towards autonomous driving is transforming the principles of industry, government, and end users alike. Edited excerpts follow.
Give me a little background on the CVTA.
McCORMICK: Back in 2000, the automakers wanted to create a competitive consortia to come to a consensus on how devices would communicate inside the vehicle, because if you were a tier 1 you made a completely different radio network connection for a Toyota than you did for GM, and it was driving cost and was all non-differentiating for the consumer. So they created an organization called the Automotive Multimedia Interface Collaboration, or AMIC. For antitrust reasons, one automaker can’t be in charge of the other. They brought me in as Executive Director of that organization, and over the next four years we developed about 3,800 pages of specifications: the SAE common message set came out of that group; the Bluetooth Hands-Free Profile; and the In-Vehicle Network Architecture specification were all written by that body. And it was not just the automakers; besides 12 automakers it was 50 global suppliers.
In 2004 I sat down with the automakers and said, “Now my next task is to move all of those specs into world standards organizations such as the ISO, SAE, etc., and, now that you have a consensus on how to communicate inside the vehicle, you need to work on communicating vehicle-to-vehicle (V2V) or vehicle-to-infrastructure (V2I) because if you don’t it’s going to be taken over by handsets and laptops.” Then I incorporated a new entity called the Vehicle Infrastructure Integration (VII) Consortium, and we wrote a $54 million cooperative agreement between the federal government and the automakers to investigate dedicated short-range communications (DSRC) technology and the 5.9 GHz spectrum possibilities for V2X communications. After a while they realized they didn’t understand how to work with the entire ecosystem because they’d primarily dealt with tiers, so in 2005 they asked if I would consider creating a new kind of trade association – one that multiple different industries could participate in. I then incorporated the CVTA, the founding board of which consisted of members of 12 different industries. So, besides having your typical Motorola and Delphi and Navtech, it also had an Intel and a Sprint and a Cisco; now there are 17 different industries that are involved.
Now we execute on a lot of things – we had a $3.5 million Michigan Economic Development Corporation (MEDC) program with another $8 million in contribution to create the Connected Vehicle Proving Center; we brought the insurers into this space; and eventually we helped bring mobile into this space.
How is the V2X communications landscape shaping up, and in particular, where is DSRC technology today?
McCORMICK: Yes. The definition of a “connected vehicle” 10 years ago was bi-directional communication through any communication protocol or frequency. So, as long as it’s bi-directional, DSRC was important for one aspect of it; cellular was important for another aspect; other flavors of Wi-Fi, Bluetooth, and satellite are all important aspects of this space. We have a number of entities, some of whom are on the board, that focus entirely on DSRC.
DSRC has been used for years – it’s used to communicate with Patriot missile systems – and it’s a very securable protocol. That was part of the $50 million that went into the VII Consortium, to discover how to do a security credentialing management system, how we would have these certificates in the car, how to use the control channels, etc. DSRC is a very unique kind of spectrum because it’s 5.9 GHz so it doesn’t go very far – by placing an antenna on top of a bridge we’ve gotten it to go two miles only because there was no interference, but generally you’re talking about 300 feet.
Initially, a decade ago, people were thinking that we would do everything with DSRC because the spectrum was free and because it has the lowest latency – 5G might compete with that when eventually it gets out there, but for the foreseeable future, DSRC is the lowest latency and it’s the easiest to package up into message sets because we know what those message sets are. The issue with DSRC comes when you start talking about communicating to the infrastructure. In the beginning everyone was saying, “Well, I guess the government will have to have to put up $10 billion worth of infrastructure and put up another tower every 300 feet.” Even back then we were thinking, “Wait a minute, guys.”
If I tell the Department of Transportation (DoT) that there’s a pothole out here, they’re not going to get to it in real time. They’ll get to it, but they don’t need the information like your car does if the car in front of you has its emergency brake lights on or is coming at you or is on black ice – the things that will takes us into the autonomous vehicle in a decade or so. So the issue is now, “Wait a minute, maybe we only use DSRC vehicle-to-vehicle,” and that’s also partially a deployment issue. If you’re doing DSRC in a V2V context it can be embedded in or deployed aftermarket, but when you’re talking to the infrastructure, if I can use 4G or LTE, that signal can hop vehicle-to-vehicle, and it doesn’t even have to go the vehicle equipment. It can go to an app on your phone that can relay or produce information that you need about road traffic or weather conditions to the infrastructure.
We only replace 7 percent of our cars each year because we’re keeping them on average 11 years. I just bought a car earlier this year, so I’m not going to get another one for another decade, and by then the technology will have evolved. But I’ll get a new phone in about 9 months. So, if my phone is able to receive this information and tell me not that the guy in front of me is on black ice, but the car a mile ahead on the road I’m taking is on black ice, that’s just as good. And, now I can accelerate deployment for the existing fleet and have that safety benefit carry forward. So we can’t look at things linearly and try to use DSRC for everything or Wi-Fi for everything, because the answer is that each protocol has a value and capability for a specific environment based on your travel vector – where you are; where you’re going; how soon you need the information; and, if you can’t get it in real time, how you can get it a little earlier so you, and not the car, are informed so that you can be prepared.
What are some of the V2X challenges facing the industry right now?
McCORMICK: The bigger issues technically are the security aspects. There’s a $10 million federal program with the automakers to secure and harden connected electronic control systems in the car, and it’s not just a question of a malicious threat getting into one of these portals because the reality is that’s much less of a concern than software collisions. The Boeing Dreamliner has 4 million lines of code; the latest luxury vehicles have 20 million lines of code. There’s no way automakers could possibly test for all the software collisions. Microsoft couldn’t for Windows 8 – there were 100,000 known bugs when it was released – knowing that they would get fixed over time. Well, that’s a whole different beast if you’re talking about an embedded system than when you’re talking about what goes on in your laptop or your phone.
Also, most of the major changes in the car are going to be electronic and 60 percent of people are now not buying a car if it doesn’t have the tech they want in it. So, where before the automakers themselves would decide on a common denominator that everyone wanted and put that in everything, like with satellite radio, now they’re having to say, “What is it that you want.” Somebody may want sports scores whereas somebody else may want stock quotes where somebody else may want their tunes. So the killer app is now essentially a browser to get what I want rather than what somebody else wants to give me.
That’s a fundamental change because historically the quality of service is what drives why people buy and stay with a particular brand. Now the automakers have extended their quality of service to include something they have no control over, which is the quality of the communication protocol – I’m in an urban canyon, does it work?; I’m in the middle of Oklahoma and I don’t have cell service, what am I going to do? So all of a sudden you have to be more conscientious as an OEM or a supplier or even as an aftermarket producer because people are going to gauge your product on things that you may not have control over.
How do you see the government factoring into V2X rollouts?
McCORMICK: I testified to both the Senate and House Transportation Subcommittees last year because under the President’s Jobs Act they wanted to allow unlicensed devices to utilize the 2.0 GHz, 5.0 GHz, and 5.9 GHz spectrum, and everybody from AAA to ITS America to the CVTA went to Washington D.C. and said, “Look, you need to protect this spectrum.”
The FCC had done tests that said that an unlicensed device properly structured can recognize that a licensed device is on this spectrum relinquish the use of it. I said, “That’s great in a lab.” The problem is that if you have 200 vehicles at an intersection, what you have is all these cars grabbing and releasing the spectrum, and that’s what’s known as a denial of service (DoS) attack. At the conclusion of that, the FCC agreed to do further testing and to not let unlicensed devices onto 5.9 GHz. We’re okay if they can say it will work, but somebody’s got to go do the work of testing it to make sure it does work, and it’s got to be done like they did the Ann Arbor pilot with 3,000 vehicles. It can’t be, “Oh, we tested it with two boxes in a lab.”
What are the objectives of the CVTA right now?
McCORMICK: Because I’ve been long in this space and chair a lot of conferences and we support a lot of events, what I’ve seen over the last few years is a lot of conferences have evolved into using sponsored speakers. So you’ll get a company that does a very niche aspect of the environment getting up there to talk about what they see for Wi-Fi or cellular or a small part of the ecosystem. And when I do the surveys of these conferences, for 50-60 percent of attendees it’s their first conference, year over year, which tells you two things: they’re sending people to feel out the environment and they’re not getting a broad understanding of what the spectrum is.
We’ve created the Connected Vehicle Professional Credentialing program. We’re creating a course, which would essentially take a year if you took all 16 modules, that talks about here are the functions; here are the communications protocols, what they’re used for, and what the potential is; here’s the difference between 3G, 4G, LTE-Advanced, and 5G; and then here are what the infrastructure needs are for embedded versus aftermarket. The purpose is to give all the industries out there an opportunity to get a foundational understanding of V2X because you don’t get that at conferences anymore and there’s no place to go read it. All of the major telecoms are interested, most of the major tier 1s have said they want to be involved. If you’re an automaker, you can’t take one of your senior people and have him spend time educating a new hire. It will be something that has to be updated every year as the world grows, but that’s really what we’re concentrating on right now.
What are your projections for V2X and autonomous driving over the next 5-10 years?
McCORMICK: We forecast, and we’ve talked to SAE and they agree, 500,000 jobs coming up in this space in the next four years alone. As insurers, telecoms, and electronics entities turn their attention to V2X and then the automated driving space, we fundamentally feel that we’re at a tipping point, and that the changes in this space could be equivalent to the economic changes that occurred with the advent of broad public web access. It takes a long time to get to that point, but I think now we’re at the point where we’re going to start to see the convergence of the Internet of Services, the Internet of Things, connectivity, and autonomy that we have to have to go forward.
Connected Vehicle Trade Association (CVTA)