I think this is absolutely brilliant and too long in coming. I agree that the addition of it as an option will likely drive consumers to want it (think Consumer Reports, "The Safest Cars."). It will also be great for new drivers who are still working out the nuances of driving.
I am not sure I would call it brilliant...it sounds to me like pushing the technology down the throats of consumers...I personally have no interests in using it, or even more importantly paying for it...and if most people are like me the value of car to car communication will be pretty slow...and what if they speak different languages! Kris
Ha, ha, Kris, "Speaking different languages!"I wouldn't go that far and be so pessimistic.
The technology has been tested for many years in Michigan. It's known to work. The issue is that this has been such a long process that some of us are worried that the window of opportunity to really leverage DSRC might be closing...
The testbed in Michigan does not include the use of other wireless technologies such as a driver's smartphones. Considering the progress of technologies these days, DSRC, if watied too long, could become a thing of the past.
Junko, I live in Canada and drive to US fairly frequently...I would imagine that Mexico-US border is even more busy...so different "language" (or more likely car to car communication standards) is a remote possibility...after all you guys stick to miles and we use km, hundreds of years has passed and this simple issue has not been resolved! And now you want my car to talk to your car!! Imagine confusion if my calculates everything in meters and yours in inches...Kris
Well, you know, as long as there are still drivers behind the wheel, if V2V is incompatible or non-existent in a neighboring country, no big deal. You simply won't have the benefits of it.
Everyone should look up online videos, taken from cameras inside cars, of real-world traffic accidents. Some are truly spectacular. And there's almost nothing a human driver can do, all too often, to avoid the crash. V2V would prevent many of these, by alerting the driver of this erractic maniac about to cross over in your path. Really. Take a look.
There are different ways this could pan out. For all who have the libertarian streak, where government involvement automatically disqualifies the entire subject matter, the IEEE and IETF have done a fantastic job at enabling what I'm doing right now, with NO government involvement. So the auto companies and universtities can certainly pull this off on their own.
I'm convinced that in a few decades, people will be amazed that we ever drove without these new measures, using only our own eyes and ears. It will seem incredibly primitive and beyond reckless.
As to V2I, as discussed several times in other articles, that's a whole 'nother matter. I think it's essential for full automation, but V2V could certainly improve safety in the meantime.
Oh, and Kris, my USan PC seems to have no trouble communicating with UKan UBM, or reading your Canadian post!!
Bert, I can read your post too...hurray!, US English to Canadian English translation is working well ;-)...the reason is that Internet was invented and build quickly around the globe...other communication standards are different story...and I wil mention again km to mile translation, major engineering projects (space rockets, missiles, cars) failed because someone forgot to convert to units...I really don't want to have more technology in my car pls, I just to want to drive! Kris
Bert, I can read your post too...hurray!, US English to Canadian English translation is working well ;-)...the reason is that Internet was invented and build quickly around the globe...
I agree that this is not always the case, but the reason that devices can interoperate over the Internet is that those who develop the protocols have so far seen the advantage of doing it this way. Enlightened self-interest.
Where I work, we used to have a complicated scheme of Softswitch gateways, to allow our IP, SNA, IPX, and DECnet networks in the company to interoperate. It didn't work very well, because naturally, it was practically impossible to make all the features of each protocol translate correctly into an equivalent feature of another protocol.
So, the auto industry can take it upon itself to follow the example of the IETF, or actually participate and use the IETF to ensure interoperability.
Different units of measure are no problem. Software can easily convert between them. All you need there is to identify the measurement standard along with the measurement, when you write the message standard. Or alternatively, the message frames can be mandated to always use the same system, say the metric system, and then any local display of the information can translate to the user's preference. (My car already allows this, btw, for the instrument panel.)
Analog radio, old-style vinyl records, and even CDs, have always interoperated globally. It's not impossible to achieve this. If there's a will, that is.
I agree Bert this is posisble to accomplish...and km to miles conversion is so dumb simple that I am still amazed that multi-million dollar programs trip over this (including US rocket launch)...but car is a car, I want 100% probablity that it will drive, not 99% like my PC working or not 95% that WiFi will be working (I was visiting a major US University recently and my laptop would refuse to read emails despite having correct PHY connectivity, probalem was at the MAC or higher level of hierarchy)...and I don't want my car operation being dependent on talking to some other car...Kris
@BrainiacVI, exactly. The auto industry is famous for dragging its heels.
Aside from seat belts, there is also an issue with backup cameras. The idea was to equip all new cars with backup camera so that drivers can see what's behind their car. It was almost mandated last year, but it fell through. Here's a clip from Auotmotive News.
DOT proposed rules in 2010 that would have required backup cameras in all new cars and light trucks. Final rules were delayed multiple times after automakers and White House officials raised concerns over costs. Before leaving office this year, Transportation Secretary Ray LaHood set a new goal of completing the standards by 2015.
car to car talk is the futuristic feature and soon would become quite useful. The standards used and regulation is something to watch for. Also there are so many car manufacturer would this communication be independant of mfr. or each of them would have their proprietary.
As the article says, car-to-car will not assist with impaired drivers. One of the biggest causes of accidents where I live seems to still be impaired drivers. I wish there was some way to prevent a car from starting if a driver is under the influence or too tired. True, the passenger could "take the test" but I think it might hit many people with the truth that they are really *not* "OK to drive."
All they need to do is enforce impaired driving laws more rigorously, and take licences away from people who have demonstrated that they aren't responsible enough to be trusted with one. This would have the side-benefit of increasing demand for public transportation, which in turn improves its quality, so people don't need to drive as much. Plus, you get the health benefits of having a population which walks more (to get to and from public transit), saving national health care costs. This is IMO a big reason why Europeans are healthier than Americans.
Now if people want to rob me and/or steal my car, all they have to do is spoof broken down cars ahead of me so my car stops itself, or trigger the kill switch.
The problem with cars is not that they are too simple and need myriad more failure modes. Can you imagine the chaos with all cars out there running different versions of the V2V software, and the chaos when 50% of them don't start some morning because they received an automatic update overnight that bricked their computers?
Can you imagine the chaos with all cars out there running different versions of the V2V software, and the chaos when 50% of them don't start some morning because they received an automatic update overnight that bricked their computers?
In my opinion, making V2V systems legal and mandatory will make drivers more dependent on such systems and less attentive in their driving; and blaming the "system malfunction" whenever some accident happens, thereby claiming damages from insurance companies as well as from the "system" manufacturers.
Such systems are helpful only if they prevent the errant drivers from breaking the traffic rules and posing hazards to other drivers . So such systems should prevent a car to cross the red signal, force it to stop wherever there is a "STOP" sign., regulate the car speed automatically as per the speed limit zones and so on.
Just alerting the driver of a possible accident situation won't help!
The problem I see with this is how will you assure even adoption. Until a critical mass of people has this technology, it is basically useless. How do you get the numbers of people driving these kinds of cars up there to begin with?
@zewde, obviously, this won't happen overnight. But either a mandate (needs rulemaking) or an industry auotmotive safety rating (safety stars?) could help propel the penetration of cars with DSRC technology.
I agree Zewde...unless government decides to mandate this technology it will be useless...hopefully there is enough push back against any attempts to legislate this...I don't want my car to be networked to anything!...the givernment should focus on delivering something useful like public transportation, buses are much more efficient than cars and they can network them as much as they want, I won't be behind the weel...Kris
I agree with the decentralized nature of the V2V discussed here. It is very tempting to try to centralize this, similar to centralized Air Traffic Control. As the article says, the latencies would be a real problem. That being said, there are still a lot of details to be worked out. Presumably the communications protocol would be self-describing and versioned, which would eliminate the "inches versus meters" problem mentioned earlier. If it included identification information and vehicle information (specifically speed) then police radar would become unnecessary (there's a selling point for you...). It would become very easy to put fixed receivers in place to track at least how many cars pass a particular point and potentially exactly who they are (another boon for the police). My first thought was that a decentralized implementation would have fewer privacy concerns, but that lasted about two minutes.
NASA's Orion Flight Software Production Systems Manager Darrel G. Raines joins Planet Analog Editor Steve Taranovich and Embedded.com Editor Max Maxfield to talk about embedded flight software used on the Mars on EE Times Radio. Live radio show and live chat. Get your questions ready.