Stories
Slash Boxes
Comments
Search

News for nerds, stuff that matters

Slashdot Log In

Log In

Create Account  |  Retrieve Password

Peer to Peer Networking for Road Traffic

Posted by Zonk on Sat Mar 17, 2007 08:14 PM
from the look-for-gridguide-coming-soon-from-saeder-krupp dept.
Networking Technology
alecclews writes "The BBC is reporting on some German research to allow the exchange of information between road vehicles about travel conditions using peer to peer networking (I assume some sort of mesh). Cars or bikes experiencing problems would pass data that would ripple down the chain of vehicles behind them. 'For example, cars could spot oil on the road by combining temperature readings with wheel traction information. A wheel slipping on the road even though the temperature was not low enough for frost or ice would suggest oil or another slippery substance was present. Once a car detected this sort of danger, information about it would be generated and passed down the line of vehicles approaching the patch of oil.'"
+ -
story
This discussion has been archived. No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.

Peer to Peer Networking for Road Traffic 50 Comments More /

 Full
 Abbreviated
 Hidden
More
Loading... please wait.

  • good and bad (Score:5, Insightful)

    by mastershake_phd (1050150) on Saturday March 17 2007, @08:19PM (#18390995) Homepage
    Could prevent pileups at the least. Of course anyone with such a system could potentially be tracked.

    • Re:good and bad (Score:5, Funny)

      by cheater512 (783349) <nick@nickstallman.net> on Saturday March 17 2007, @08:22PM (#18391017) Homepage
      Even better. If your late for a meeting just fire up a laptop and tell all the cars that there has been a major crash.

      Everyone avoids it allowing you to get to your meeting in time.
        • If any single car were trusted by the rest, any jokester could cause chaos on a whim.

          On the other hand, with a convoy of friends (perhaps "haha, you opened the sexy_pix.gif___.pif attachment in your MS Outlook Ford Edition" friends)....
          • Or you could make life even easier. You and a convoy of friends could just stop. Sure it will create chaos behind you, but whats telling the computer that there isn't a herd of cattle crossing the road.

            • Re: (Score:2)

              by JavaRob (28971)
              What, are you some kind of luddite? Plus, you'd get caught pretty easily if you just blocked the traffic in the obvious way.

    • Re: (Score:1, Interesting)

      by joshier (957448)
      It's not a matter of choice, cars will be able to be tracked at some point, but I think there may be choice for a while

      What I do think though, is that I've always thought about this kind of system, whereby if there is a care accident, each car that is surrounded is alerted and the driver is made fully aware.
      As for traffic accidents in the UK, I think this system would be very welcome.. Personally, I would very much welcome a system like this ... not everyone is 100% fully aware and even when we are, we

    • Re: (Score:1, Interesting)

      by Anonymous Coward
      i would like to see this used with a heads up display on the windshield to mark the dangers (like oil or ice), thereby avoiding annoying computer voices, alarms, idiot lights that we all ignore, etc.

      probably a little further off, but i think a better system.

    • Re: (Score:1, Interesting)

      by LiquidCoooled (634315)
      Here in the UK our traffic cameras are about to be upgraded to included license plate identification.
      There are probably more cameras than cars on the road.
      Do you think they need to put something in your car to know where you are?

      • There are probably more cameras than cars on the road.

        There are roughly 6000 cameras [speedcamerasuk.com] on UK roads. Compare to 33 million [whatcar.com] cars. Stop spreading FUD.

        Do you think they need to put something in your car to know where you are?

        If you're going to implement road pricing, then yes. The alternative is installing cameras on every road in the country.

    • Re: (Score:2, Interesting)

      by Cstryon (793006)
      I could see more bad.
      A car in front if me gets the signal there is an oil slick comming up, that guy is gonna change lanes to avoid it completely. I could get the signal just after him, and so on, everyone will want to change lanes. And of course if someone behind me wants to change lanes and drive faster then everyone else, sideswipes, people getting cut off, panic. It just sounds like a bad idea all together.
      • Fortunately any decent peer-to-peer system has methods for controlling such threats.

        The need for data to "change lanes" is an extremely common occurrence in peer-to-peer networks. Modern peer-to-peer systems generally handle the situation extremely well. The fact that your computer doesn't instantly crash with thousands of requests whenever you join a gnutella network shows that the concept is well handled.

        The German system described in TFA seems designed to stop everyone instantly switching lanes, as one w

    • Re: (Score:2, Funny)

      by Anonymous Coward

      (I assume some sort of mesh)
      Really? I would think a bus topology would be more appropriate. Arr, arr
    • Could also cause pile-ups, as someone gets a warning that there's oil on the road ahead and slams on his brakes "just to be safe". The traffic patterns formed by drivers who can only see a short distance ahead can be frustrating, but at least they're fairly consistent and predictable. Introducing widespread limited clairvoyance of this sort would change that, and not necessarily for the better.

  • Great idea! (Score:5, Funny)

    by NtroP (649992) on Saturday March 17 2007, @08:24PM (#18391033)
    Now any idiot with the right cantenna can bring any highway to a crawl by inserting bogus messages into the mesh. "Look out! Slippery road. Warning, stopped traffic ahead. Pull over, emergency vehicle approaching from behind. Look, Elvis!"

    • Re: (Score:3, Interesting)

      by kalirion (728907)
      Personally, I think more people will be interested in redirecting traffic around the rounds they drive on, especially in rush our.
      • They would ban any such system that let people decide what got sent. Mostly because people would warn others about traffic cops, decreasing revenue.

        • Re:Great idea! (Score:5, Insightful)

          by NtroP (649992) on Saturday March 17 2007, @09:18PM (#18391389)

          They would ban any such system that let people decide what got sent.
          Right, 'cause this system will never be cracked and "banning" something automatically stops people from doing it, right?

    • Re:Great idea! (Score:5, Funny)

      by haakondahl (893488) on Saturday March 17 2007, @09:33PM (#18391457)
      "Soon, our vehicles will all drive themselves."

      "Soon, our vehicles will all talk to each other."

      Soon, our vehicles will all get tickets for driving while talking on the phone.

    • A similar objection (Score:5, Insightful)

      by dsanfte (443781) on Saturday March 17 2007, @09:55PM (#18391551) Journal
      We once had this idea for a global voice network. Everyone would have a number and accept calls by default, and people could talk to each other. One guy on Slashdot knew better, though. He informed us that people could call businesses with bomb threats, for example, and disrupt the economy. Adults could call children and try to abduct them. Random businesses could harass individuals with marketing calls. Loopholes abounded and there was no way to fix the system without breaking it more.

      We would have called this a telephone network, but we had to give up on it since its security was obviously so flawed. Thankfully that guy on Slashdot saved us all that wasted infrastructure money. Nothing good would have come of it anyway.

  • Four pieces of data and repeaters (Score:5, Interesting)

    by Harmonious Botch (921977) * on Saturday March 17 2007, @08:25PM (#18391047) Homepage Journal
    I've been thinking about this one on occasion for a few years now...

    The only things that need be passed along are current GPS location ( deliberately imprecise by about 20ft ), current velocity ( deliberately imprecise by about 10mph ), last 5 secs acceleration on all 3 axes and a time stamp.
    The other function that a car should do is listen to the traffic going the other way and pass on an average of what it hears. ( This averaging function is crucial. It enables velocity and location to be reported without giving up evidence of speeding.
    As an example: northbound traffic reports the four pieces of data. Southbound traffic listens to it and averages it. A minute or more later the southbound traffic repeats that to the northbound traffic who are soon to encounter the situatuion. It keeps repeating it - interspersed with other data about other locations - with decreasing frequency as it gets further away.
    • It is indeed interesting. Under very controlled situations it might be useful. The trouble is that the standard highway is not a very controlled situation. The mesh links would break as traffic separated, or by cars without the networking installed.

      GPS coordinates of the message origination vehicle would be good, but estimating how fast your vehicle will arrive at the trouble spot is complicated by erratic driving of any vehicle between you and the trouble spot. Within seconds (the presumed latency of the m

      • Within seconds (the presumed latency of the mesh) the trouble spot could move 100s or 1000s of feet closer to you.

        That is the point of having the average repeated. As you are driving north, your car would hear about a problem spot numerous times from southbound traffic. Your car can take these reports and conclude that the trouble spot is moving.

        God help you if you are following someone that thinks it would be interesting to run under the back of a large truck at 100mph, or is trying to commit suicide.

        That is compensated for by averaging. If there is an unsafe driver, a suicidal driver, or even a deliberate lying broadcaster, he gets averaged out. ( Actually the process is a tad more complex. A mode value calculation removes the freaks. Compare that to the averag

        • Averaging the data of a dozen cars in ten minutes (late at night, low traffic area) will give each car considerably greater weight (and more overall vulnerability to issues like GP posted). Unless there is at least moderate traffic to provide sufficient population, I don't see averaging as a reliable/safe solution.
          • Thanks, but this is more complex. The Byzantine Generals problem deals with data that is discrete, and often binary. The traffic problem deals with data that is continuous.
            In other words, in the Byzantine Generals problem, if A != B then B is a different class of data form A. In the traffic problem, A and B may be in the same class - and treated as (A+B)/2 - or they may be different as in the generals problem.

    • Re: (Score:3, Informative)

      by daeg (828071)
      Much of this information can already be obtained from you cell phone company. No need to have it based on a separate device.

      I believe a few cities are working on implementing this, Tampa being one of them.
      • I hope everybody's cell phone has a 3-axis accelerometer.

        Deducing acceleration from location has one of two problems: either it is precise enough to constitute proof of speeding ( which will lead to deliberate non-participation ) or it is not precise enough for that in which case accurate values for acceleration cannot be calculated. As my OP says, speed data needs to be deliberately fuzzy. But acceleration values must be very precise.

      • Re: (Score:2)

        by caluml (551744)
        Look up Java J2ME JSR 179 - it's the location API that's supported on modern phones.

    • The only things that need be passed along are current GPS location ( deliberately imprecise by about 20ft )

      GPS hasn't been deliberately imprecise since 2000, and systems like WAAS [wikipedia.org] and EGNOS [wikipedia.org] that are fitted to most new in-car GPS systems allow you to get readings typically to within 3-4 feet. The 20 feet you quote above is the typical accuracy of plain GPS when it is not being tampered with by the military, the official spec for GPS (with deliberate imprecision) only guarantees accuracy to within 300ft.

    • Re: (Score:2)

      by caluml (551744)
      The only things that need be passed along are current GPS location ( deliberately imprecise by about 20ft ), current velocity ( deliberately imprecise by about 10mph ), last 5 secs acceleration on all 3 axes and a time stamp.

      That's exactly what I do [calum.org]. Except I round off the GPS lon/lat to a certain amount of decimal points - that basically makes it inaccurate enough not to incriminate me - and I don't bother with a timestamp - saves on bytes - I just use the time at the server when it's received.

  • IPv6 (Score:3, Insightful)

    by Anonymous Coward on Saturday March 17 2007, @08:31PM (#18391085)
    It's application like this that should make the need for more IP address space obvious. There are other ways, but nothing is so elegantly simple as handling your car's computer as just another device on the network, addressable on the Internet when possible. In the not-too-distant future, it should be possible to access your car's performance data without buying expensive equipment from the manufacturer.

    • Re:IPv6 (Score:4, Insightful)

      by JPriest (547211) on Saturday March 17 2007, @08:52PM (#18391239) Homepage
      In the not-too-distant future, it should be possible to access your car's performance data without buying expensive equipment from the manufacturer.



      So you think IPv6 suddenly means auto manufacturers will stop being so proprietary?

  • look for abuse potential before implementation (Score:1, Insightful)

    by Anonymous Coward
    As with all things good (and automated), we should be looking for abuse potential before implementation. For instance, could the system be hacked to:

    a) provide erroneous information (general nuisance)
    b) provide erroneous information to cause intentional lockup (i.e. a special-interests group publicity/demonstration)
    c) provide erroneous information so vehicles are forced to not follow in the footsteps of a vehicle (black cars/helicopters that dont want witnesses for some secret CIA operation, yadda yadda)
    d)

    • I'd be more concerned about the potential for abuse by the already abusive and power-hungry government. They are already installing black boxes in automobiles, I see this system ratting out to your friendly highway cruiser that you were speeding a few kms back. [/tinfoil]

      Also they would have to have pretty strong and resilient wireless. Right now its a Major PITA to use my Bluetooth stereo headphones to listen music in the mobile for long periods without the connection breaking, and the handset is in my po

  • Up with which I will not put (Score:4, Funny)

    by malia8888 (646496) on Saturday March 17 2007, @08:44PM (#18391181)
    When the motorbike comes after to the point of danger, information has been spread out by wireless network and the danger will be propagated to the driver in the motorbike Dr Anselm Blocher

    I read this three times and thought I was retarded. :P
  • 1. Commit Crime
    2. Jump inside get away car & drive off.
    3. When police begin approaching, connect Laptop into getaway car computer system & insert bogus messages such that they propogate to the Police cars behind & anywhere around you. Wireless amplifiers here will be really useful. Suggested message could be "Bridge Out" which would bring every car on the road to a full STOP so you can just drive around them all.
    4. PROFIT !!!!!!!!!!!!!

  • Snow crash (Score:3, Insightful)

    by LiquidCoooled (634315) on Saturday March 17 2007, @08:51PM (#18391231) Homepage Journal
    This reminds me of snowcrash:

    Out in the world beyond his yard, there are other yards with other doggies just like him. These aren't nasty dogs. They are all his friends.

    The closest neighbor doggie is far away, farther than he can see. But he can hear this doggie bark sometimes, when a bad person approaches his yard. He can hear other neighbor doggies, too, a whole pack of them stretching off into the distance, in all directions. He belongs to a big pack of nice doggies.

    He and the other nice doggies bark whenever a stranger comes into their yard, or even near it. The stranger doesn't hear him, but all the other doggies in the pack do. If they live nearby, they get excited. They wake up and get ready to do bad things to that stranger if he should try to come into their yard.

    When a neighbor doggie barks at a stranger, pictures and sounds and smells come into his mind along with the bark. He suddenly knows what that stranger looks like. What he smells like. How he sounds. Then, if that stranger should come anywhere near his yard, he will recognize him. He will help spread the bark along to other nice doggies so that the entire pack can all be prepared to fight the stranger.

  • I could see using a system like this to relay traffic conditions (IE all the cars on highway 40 are going 5-10mph), but the example cited in the blurb is truly bizarre. Even if you could relay "There is some oil on the road at mile marker 22.5" even if you could use GPS coordinates... How are you going to specify anything besides "right where my wheel went over". In short there is no way to make this information specific enough to actually aid a trailing driver in avoiding the danger.

    If this were implemen
  • It's Eastern Standard Tribe made real.

  • The most important safety benefit... (Score:4, Interesting)

    by Anonymous Coward on Saturday March 17 2007, @09:20PM (#18391403)
    .... would come from VoIP connectivity between nearby vehicles. We wouldn't be as prone to road-rage type behavior if we could easily speak with people in adjacent cars. If I could say, "Ahem, excuse me," instead of tailgating and flashing my headlights at the idiot camping in the passing lane, both his and my blood pressure would benefit.

    A lot of dangerous/reckless driving behavior comes down to the dehumanizing nature of cars. If you see traffic as a collection of people and not anonymous metal cages, you'll be a more considerate and safer driver.

    Automatically spotting and checking for oil on the road... yeah, I guess that's cool, but it's not the most important use of this tech.
  • What happens when other people can remotely influence the performance of your automobile? There may not be a case for "remote control" (one would hope), but injecting false information into the network about dangerous traffic conditions has implications for other vehicles programmed to respond to it.
    • Once our cars are on autopilot -- DARPA is working on it through its neat races -- do you suppose we'll have administrator access to them, or will someone else?

  • It's Been Around for a While (Score:4, Informative)

    by StormyMonday (163372) on Saturday March 17 2007, @10:30PM (#18391723) Homepage
    The network part is very much Old News; it's called Dedicated Short-Range Communications (DSRC) [wikipedia.org]. It's been around for several years, and there are a number of standards committees working on it.

    Last I heard, a year or so ago, there was a limited rollout planned for some luxury cars in the 2008 model year, with some simple car-to-roadside communications (map updates, traffic signal status, etc).

    The new part here is using AI to sort out what information to give to the driver, and how. It's obvious that if you're not careful, you'll swamp the driver in information.

    Coupla other items:
    • Spoofing: A problem. Last time I worked on it, they were looking at some digital signature tricks.
    • Privacy: A problem. Basically, every time a radio goes out of contact, it randomizes its MAC address. It'll work fine -- if it's properly implemented. Remember WEP?
    • IPv6: Yuppers. All the way.

    • The network part is very much Old News; it's called Dedicated Short-Range Communications (DSRC) [wikipedia.org].

      Please Read The Fine Subject. "Peer to Peer Networking...".

      You are referring to a network solution where vehicles talk to stationary network gear. Yes, that is old news. When I was a child 30 years ago, such systems were predicted.

      However, the article seems to consider a peer-to-peer network where nearby vehicles communicate with each other and relay information to other vehicles which are outsi

      • Re: (Score:2)

        by uradu (10768)
        This has been done before invarious ways. In the 80s VW experimented with a roadside beacon approach that communicated with an in-car display system: when the car entered a road segment leading to a serviced light, one of three LEDs on the dash simulating a traffic light lit, indicating the color of the upcoming light at the current speed. The idea was to choose a speed that kept the on-dash traffic light green, presumably within the legal speed limit. They actually had a pilot installation in a couple of c
  • I thought Emule and Bittorrent via WiMax for my car stereo was here.
  • As long as they don't start using bittorrent to get us to our destination.

    "Sorry boss. Part of me is still out on I-25. I'll be in the meeting as soon as the pieces arrive."

    H.
  • I came up with something similar about six months or a year ago which I dubbed "CopWatch."

    Basically, it uses the same principle, but every time you see a traffic cop, you press a button somewhere in your car. Your car, with the use of a GPS, then beacons the location of the police car. Other cars then repeat the beacon, which does have a TTL value on it as well.

    To prevent false positives, there is a limit to how many reports someone could generate in a set time period, and multiple reports in the same area
  • while driving? Sorry, this existed 30 years ago. "Breaker, breaker, good buddy on channel 13, smokey under the bridge...."

          mark "I've got our CB radios here *somewhere*...."
    • .. and the rest of the EU as well.

      And make no mistakes, if the US Government thought they could get away with it, they would too.

      I'ts ostensibly to support "congestion reduction" through road charging. But there are other ways to implement that that don't require a GPS tracker in your car 24/7.

All trademarks and copyrights on this page are owned by their respective owners. Comments are owned by the Poster. The Rest © 1997-2010 Geeknet, Inc.