Connect with us

cars technologies

Discussion in 'Electrical Engineering' started by Jevan, Feb 12, 2004.

Scroll to continue with content
  1. Jevan

    Jevan Guest

    (Re: Article by Jim Louderback - editor - Ziff Davis Media - PCMag/eWeek)


    See below. Not too bad... I proposed this in discussion (not written in the
    thesis though, just said it to my supervisor Dr Ken Taylor) when I did my
    4th year Engineering project. Personally though I took it a step further,
    saying that they'll have it automatically take over the car driving
    completely, if need by to avoid an accident - steering and all. After all, I
    was doing the speed control for automation of the landcruiser project, but
    why not also control the steering (we planned to, anyway!). You'd need damn
    good software to interpret pedestrians, road signs, etc better than a human
    can - it'd have to, to be able to take over from us safely.

    However it can calculate the directions and so on of all other drivers in
    their cars, and using engineering knowledge built in, can know exactly which
    way to turn or brake or accelerate, in order to avoid and/or minimize any
    impact or accident situation. After all, we have excellent knowledge of
    collisions, momentum, friction, engineering-dynamics, etc also we can make
    prediction of potential ability to turn (given the computer can read speed
    and direction from the cars instruments) or what *each* particular car can
    conceivably do by way of manouvring (you'd have it to do a periodical
    calibration per car or base it on age of tires and type, car model/make/age
    etc) so why not use it to know what exactly the car is capable of, and use
    this in automatic accident avoidance.

    Taking it a step further, once all cars are networked (they can all have
    little infra red sensor and transmittors on the bumper bars perhaps, for
    example, or short range (50 metre max) radio transmittors/receivers), then
    they can in fact make a COLLECTIVE decision (intelligent) all by computer,
    on a networked basis, about which way each cars automatic control system
    will react in order to minimize and hopefully also always avoid an
    accident - so that accidnets will in fact be a thing of the past and now
    almost impossible to happen (certainly by human cause impossible nayway - if
    your wheel suddenly falls off causing your car to hurtle toward another car,
    then you'd still have the network making a 'best possible comprimise' in
    order to minimize any accident situation for all drivers, for example the
    car would try and compensate using steering, braking, whatever, for the
    situation ,to attempt to safely stop, and nearby cars would automatically
    steer away from the onciming path, in the intelligent context of KNOWING
    *exactly* what the car is actually about to do in the context of its
    situation, since the cars computers controllers are now communicating with
    each other (which is what wil happen automatically if a potential accident
    situation is ever detected by any car on the road).

    I think a computer system is potentially far outweigh anything a human could
    do.

    But not only in accident avoidance - also in driving. After all, if you feel
    like it you can drive, but if you want to talk to the kids, just switch it
    to automatic and let the car drive you to work, after you punch in the
    destination on the map/ GPS aware/ etc.

    Or maybe just say where you want to go and it confirm it first with a street
    map or photo to make sure. After all, we now have online detail maps (is it
    zoom from satellite?) of a lot of the world, you can go on a virtual 3d tour
    to lots of places and see what it actually look like being in the street
    there (video footage?), take it a bit further and you can actually see the
    buildings and people (defence policy restricts the satellite images
    resolution unfortunately, but anyway this is the future, maybe not anymore
    then). You can check if your workplace looks open first, by looking up a
    recent (last 5 minutes?) satellite scan of that portion of the globe (if
    this actually possible to store and scan that much information of the entire
    globe and that often though..)? Aim for the ground and you'll get the
    ground, aim for halfway ot the ground and you only get halfway, so why not
    aim further (though yes I agree, do things in small steps is good).

    I think his article is just the beginning of the future, that's only the
    beginning.

    Also he talks of hybrid - fuel cell + petrol. What about solar? Why not
    hybrid fuel cell + solar + petrol, then?

    Personally I think, what if we weave in solar material into clothing, then
    you can sit in the office all day and come home all ready to plug in and
    charge up your car. Well, perhaps clothing is not quite necessary, cause you
    wouldn't get much from fluro lighting moreso only really for outdoors,

    I suppose perhaps when you hang out your clothing to dry after being washed,
    then you bring it in hours later and plug it in to charge up the car.

    After all, surely the solar cell silicon lines need not be rigid nor in a
    straight line, can you coat them in a extremely thin flexible plastic layer
    followed by a flexible nylon? protective layer (someone can figure out the
    materials to use.. has to be waterproof but not cause any annoyance when
    worn) and then weave them as a thread in amongst other threads making up the
    garment.

    I think the only issue would be how to store the energy - ie battery.
    Perhaps the "Australian washing line Mark II" can be produced which picks up
    current flows from underside of the clothing, and forwards it into charging
    a built in battery (perhaps a fuel cell, at that!? since they so efficient?
    and only water as a byproduct.. for environment friendly).

    Maybe that another one of my off the planet ideas.. but who knows... imagine
    how many pepole hang out their clothing in the world, imagine how much
    sunlight and energy would get picked up.. would it be enough to power you
    house and car for the week? or day? Or maybe with some more improvements in
    solar technology? Just an idea.. I think one solution to effectively
    increase efficiency is to cover everything we own in solar cells (hehe).

    How about bricks that are solar cells. Then the bricks plug together,
    however way the builder happens to put them, so that the whole walls and
    roof (tiles too!) are now solar cells. How about making the solar cells
    material transparent so all you see is normal looking bricks and tiles.
    track cars around you, ensuring safe separation by warning you >first, and
    then automatically braking or speeding up to avoid an >accident.

    http://story.news.yahoo.com/news?tm...0205/tc_techtues_zd/118388&cid=1739&ncid=1729

    sports and weather, a rolling internet node will help you get around
    accidents and avoid problems. A new type of wireless technology -the mesh
    network- will automatically and dynamically configure and reconfigure a
    network using all the cars in, say, a 500-foot radius. If one car spots
    trouble -an airbag deployment, or something in the road -- all the other
    cars down the line can be warned in advance.

    Entertainment: As flat color displays get cheaper, expect them to
    proliferate throughout the car. The entertainment hub - a combination movie
    server, TiVo (news - web sites) and game console - will live under the front
    seat or in the trunk of most cars. Advanced audio technology and
    noise-canceling headphones will turn every car into a personal home theater
    customized for each passenger. Most cars will be outfitted with satellite
    video and audio, helping to make the car at least as connected as the home.


    Navigation: Today's add-on systems will seem antiques compared to the
    advanced navigation systems in tomorrow's cars. High tech road sensors will
    communicate with the traffic sensors in your car to help identify - and then
    avoid -- traffic jams. And those same networks will also make it easier to
    find a gas station, hamburger joint or bathroom, just when you need it most.


    Driving Experience: Dashboards will increasingly resemble computer displays,
    with only a handful of knobs and switches. Someday voice recognition will
    advance far enough to let us speak our commands, but it will take a while to
    perfect. "Up" and "off" sound remarkably alike - especially when you're
    trying to kill the radio.


    Electronics: The roar of the road will be reduced to a murmur by special
    noise-dampening plates on windows, doors, ceilings and floor. Fiber optics
    and LEDs will replace wires and light bulbs. Mechanical controls, like the
    brake and accelerator, will be replaced by electronics. And much like
    today's airplanes, cars will bank into turns, making them more maneuverable
    and responsive.


    Don't be fooled, though. Automotive technology may have some amazing
    potential, but it needs a lot of work. As long-suffering computer users
    know, it's best to shy away from version 1.0 of any new technology. Alas, no
    one told the designers of BMW's 7 series. They built in a radical new
    controller - called the iDrive - that failed miserably. Difficult to learn,
    horrible to use, it also had a tendency to fail. Respected auto analysis
    site Auto Spies put it best, "the concept is brilliant but the software
    interface is idiotic."


    We shouldn't be surprised - because the iDrive was built on top of Microsoft
    Windows. The next iDrive version - which was rushed to market - performed
    better, but woe to all the version 1.0 users: you can't upgrade a car.

    Jim Louderback is the Editor-in-Chief for Internet sites at Ziff Davis
    Media, which runs the popular technology sites PCMag.com and eWeek.com,
    along with print magazines like PC Magazine, eWEEK, Electronic Gaming
    Monthly and Computer Gaming World. Jim's first adventure with computers
    began with playing Star Trek during high-school on a PDP-11. Since then he's
    developed applications and installed networks for many Fortune 500
    companies. For the last 12 years he's been reporting on the technology
    industry in print, radio, television and the Web.



    Cool Car Gadgets:
    - Clarion VRX935VD Car DVD Player
    - Delphi XM SKYfi Radio Portable XM
    - Garmin StreetPilot 2610 Automotive Navigation

    More Tech Tuesday for February 10, 2004:
    - Cars Go High Tech
    - Hack Your Car
    - The Best High Tech Cars
     
  2. Control could be mediated through a signal strip in the highways. Only
    those cars that respond positively to input signals would be permitted
    to move on the highways, so free moving cars would have to be excluded.

    Railways have been running (essentially) diverless trains for decades.
    For example, the drivers on the Washington D.C. Metro are there as
    window dressing for the public. They do not do squat. The computers
    drive the trains. Digital vision devices are not required for this
    system to work.

    Bob Kolker
     
  3. Guest

    <Shudder>

    Mati Meron | "When you argue with a fool,
    | chances are he is doing just the same"
     
  4. Guest

     
  5. You're basically talking about an Expert System inserted
    in between the manual controls and the roadway. Sounds good
    to me; I've always wanted a car at leasr as smart as a fish.
    Fish manage to perform complicated maneuvers en masse
    without colliding. I'd prefer to be able to just "herd" the
    car ("go that way, and handle the details") if I felt even
    slightly distracted (or distractable) instead of having to
    focus all my attention on the driving process fulltime. Now,
    which "expert" do you model the system after, some Grand
    Prix speed demon, or your spinster aunt Betsy? Or maybe a
    rainbow trout?

    Of course, if the thing starts losing competence, I want
    full manual control on demand.

    I can't wait for the inevitable Windoze- and Apple-based
    autodrivers' incompatibilities. They used to call cars
    "buggies"; that could come back with new meaning.
    If you break a job into small enough tasks, and have fast
    enough processors, decent hardware interface, etc, I agree.
    _IF_.

    Electric. But the least stupid idea I've ever heard is to
    tax every driver to support a system involving power
    conduits in the roadway itself, and fairly short-term
    storage capacity in the car to handle system glitches. If
    you are licensed, you use it. If you aren't, you don't need it.

    For longer interstate stretches it wouldn't be
    economical, and you'd need a special endorsement (competency
    test-driven) and independent power.

    It'll take some major infrastructure rearranging, but
    that's coming slowly anyway.

    Works for me. Get together with Feerguy.

    Mark L. Fergerson
     
  6. Spaced closer together? I don't think so. DO you have a link to share?
     
  7. Probably the platooned cars were going the speed limit and properly spaced
    by the "three second" rule. Real traffic routinely spaces itself so
    closely that there's no room for a lane change and no room to stop,
    although there always should be.
     
  8. Ian Stirling

    Ian Stirling Guest

    It's probably limited by policy reasons.
    The technology can generally cope with seperations of well under a meter.
    You may be talking about over 10 cars/second, for a convoy going at speed.

    You've got to consider problems of course.

    For identical cars, if the cars are autocontrolled, whatever one car
    encounters, the next one will too.

    If the first car got round a bend, then all the rest will.
    There are a very few faults that would cause problems.

    Any uncontrolled braking of any vehicle in the platoon will have minimal
    effect, as it can brake no harder than the vehicle behind it.

    There are problems.
     
  9. Noonan

    Noonan Guest

    I'd call it a draw.
     
  10. Jevan

    Jevan Guest

    Yes well, everyone knows that greater distance between cars is safer - if
    something go wrong with one car, gives the others more time to react.

    I don't think you'd want the cars closer together and driving faster, just
    because its automatically computer controlled. I think it more a case of,
    you could sit back and watch TV that drop down onto in front of the
    dashboard, listen to music, close the windows even (you don't have to look
    out any more if you don't want to know about the traffic jam up ahead that
    could still exist after all, its the humans that all decide to leave at the
    same time and during rush hour), catch up on some work on the way there.

    Maybe they should stagger working hours, to finish between 3pm and 7pm
    instead of all finishing at 5pm.... but it is convenient to all finish at
    5pm because then you get to be in the office during the same hours.

    Actually I think communication technology is going to make offices no longer
    required (after all, I.T. contractors work from home sometimes for example,
    over the network). But we'll still like offices, or common meeting places,
    because its good to get together in person. Just you won't have to do it as
    often.. maybe once a week.

    So then the load on the roads might decrease somewhat.. in theory.. enabling
    us to sustain a greater population that still seems to keep going up, but
    without impossible levels of road infrastructure.

    How are we going to do it in the year 3000.. ? what's the projections I
    wonder..

    Maybe we'll all just need to invent amazing flying vehicles so we can occupy
    all the available airspace as wel as groundspace, like in scifi..

    JP
     
  11. Jevan

    Jevan Guest


    I wonder what went wrong in the train collision situations. Seen on the news
    in the last few years - incidents of one train immobile while second train
    runs into it.

    In one situation, the people on the immobile train could see the moving
    train coming toward it from afar, but couldn't understand why it kept going
    toward them. Eventually the became alarmed and stood up (but hadn't jumped
    off the train). Also this worsened injuries because then you had standing
    people during the collision (apparently worse than if they were sitting). If
    they had known it was going to collide, they actually had plenty of time to
    jump off their immobile train, but they hadn't thought that this would
    happen (I suppose once it got too close they must have known).
    But you wonder because presumably then, maybe not all countries track all of
    their trains positions. One would assume that a train in the system you
    described, would still have its position tracked, even if the "ATO Start"
    button had not yet been pressed. Say for example if something failed on the
    railway track causing it to stop and then it deactivated overnight, and
    another train came along then the next day.

    Maybe there could be a simple & inexpensive technology you can sell to other
    countries, to install in trains to avoid such situation of a collision of
    trains. Even a simple distance sensor (laser position sensor) or something
    that point out in front & behind the train, to stop the train if it see
    something ahead in front (with sufficient time to stop) or warn if something
    behind it getting too close or accelerating toward it from afar. Has to
    handle bends in the tracks, but presumably curvature must be limited anyway
    in order for the train to be able to attain speed?...

    But as for cars - yes, well, if something is possible (which it is), someone
    will do it (fully computer control of car, with sensing/ awareness for
    things like road signs/ pedestrians/ etc). Indeed it is true now cheap video
    cameras are available and the only real expense is development cost, but for
    research organisations they may already have need for projects anyway (and
    government funded, presumably). I suppose, major car manufacturer could fund
    such projects. I agree though, got to have a good implementation. No
    bugs!! - or at least, must be failsafe.

    Causing the car to "stop" is not really enough, if something goes worng, for
    it to be failsafe. Well, if the car in front brakes, because their computer
    video gets clogged up with dirt and it can't see properly, that could cause
    an accident too. I suppose you'd also need redundant mechanisms that
    automatically take over if need be (for example if you have three sensors,
    if one doesn't match the other two then the nonmatching one gets shut down).
    Aircraft already have redundant systems, apparently (for example).

    JP
     
  12. Jevan

    Jevan Guest

    In another train collision incident I read about - again one train was
    immobile and second ran into it. (not talking about USA here for this
    example, can't remember where it was but anyway).

    In this one, the train suddenly stopped and passengers thought it must have
    got to the station and so they stood up, then the collision occurred which
    made it worse that they'd stood up (its another example of passengers
    standing prior to the collision).

    I wonder why it stopped. Perhaps a failsafe mechanism activated? But more to
    the point, why then didn't the second train stop, which then ran into it.
    Presumably another train on the line, would stop, if it detects itself
    getting too close to another train which had stopped due to failsafe
    activating?

    JP
     
  13. Jevan

    Jevan Guest

    In a third train accident situation i read about.. a train went off the
    tracks, and then all the carriages behind it therefore got into the accidnt
    as well cause they were caused to leave the tracks too.

    The problem with rail is that if something goes wrong its really designed to
    stay on the tracks, you can't just drive the train around the problem.

    I thought what if they'd mounted wheels on the train.. that is, so in the
    event of it dismounting from the track, wheels automatically come into
    action which take the weight of the carriages, allowing it to drive onward
    now off the tracks, while braking, and also the carriages separate
    automatically so they don't pull each other into worse situations, maybe
    they have some kind of elasticity at their ends so if their ends collide
    then it reduces the effect.

    Actually, another one of my ideas I had (back in 2000) was to do with
    accident impact minimization. The thought then begins with, that since you
    need to minimize both external-to-car and internal-to-body momentum changes.
    That is, you could design a car that won't shatter or break during impact,
    and will just bounce off things and stop, unharmed, and you can strap
    yourself in to the car so you don't hit anything or undergo any collisions
    within the car (eg whiplast etc) but the acceleration changes would kill you
    even if the car could handle it and even if you were strapped in so no part
    of your body moves with respect to the car during the collision - since
    internal organs of your body still can move during acceleration.

    Therefore the idea comprised segments as follows:

    1) external minimalization: designed cars so on impact, maximum bounce
    effect and do not have it compress/break any more (as we at present have it
    break slowly over longest timespan during headon collisoin for example, to
    minimize impact forces by maximizing the time). Instead, we could design it
    so the car external doesn't break and design it so it bounces, with maximum
    energy transfer into the 'bounce' so that you now have elastic collisoins
    with anything it runs into. EG if it run into a wall it would bounce off,
    whether head on, side on, etc, maybe make the car round or give it an
    exterior shell buffer layer. Therefore this gets rid of external damage to
    car itself during collision

    2) Have to design car so it can internally handle the acceleration changes
    inherent in the 'bounce' effect

    3) Have to isolate people inside, so they do not feel effect of acceleration
    changes. For this one, I proposed contain the person in a cylindrical
    rotating enclosure. This is designed so that normally it doesn't rotate, but
    if there's an acceleration change too great, then it releases the lock,
    allowing it to translate acceleration changes into rotational energy - the
    idea is that, the same amount of energy change, over a great enough radius
    of sphere, will no longer be any noticeable acceleration change - a large
    increase in energy cause only a small change in velocity near the edge, if a
    big enough rotatable sphere. Like I got the idea from the gyroscope - you
    have a rotating wheel, and if you turn the wheel left (while it still rotate
    at same speed), the platform you are on then rotates the other way.

    So the idea there is to design some kind of mechanical system which converts
    the energy of a significant change-of-acceleration of the car, into only a
    small change in velocity & acceleration actually being felt by the person.
    After all, is this actually possible to do? It seems to be, intuitively
    possible... after all, making use of some kind of mass to convert the energy
    to somewhere else, or something?

    I don't know if that's really enough..

    Maybe someone else might think up another design completely - the idea is
    basically how you can have collisions happen (at any speed eg set the
    fastest possible a car could ever do in theory, while maintain traction and
    driveable ? 120km/h? or 300km/h or more if apply to racing events as well?)
    then design the system so the cars can bounce and no acceleration effect on
    the people inside - if this is even possible..

    Maybe the system can have an internal component that momentarily accelerate
    in the opposite direction to whatever change occurred (into store into a
    spring eg) then the spring releas ethe energy into some other direction in
    which it will release much slower or only small effect? Bearing in mind very
    small timeframes and significant accelerations talking about...

    If you have something travelling forward at 60km/h, stopping it linearly is
    going to be a pretty great deceleration (relatively speaking). But if you
    turn it into a curved path while it decelerate, that would be lesser
    deceleration. That's where the rotational inner part idea came from.

    I don't know if it a bogus idea.... maybe its just not going to have any
    effect on deceleration??

    Ideas anyone? ...


    JP
     
  14. Jevan

    Jevan Guest

    Basically he says the question is whether it worthwhile of budget to equip
    trains with ability to sense obstacles in front/ behind them (diminishing
    returns by creating more and more new failsafe mechanisms), perhaps cheaper
    to rely on observer (train driver) in train to see them. Personally I wonder
    if he'd notice, if the rate is one such accident occurred in a drivers
    entire career. Like waiting for the ball to fall into the water
    (electromagnet deactivates when battery runs out, therefore unknown by
    participants when exactly this happens) during a TV series (part of a
    challenge), when it happened they weren't really expecting it or fully ready
    for it (or even know exactly what to do in their case).

    He says that people didn't jump off the train before the collision that
    injured many of them, because you'd be embarrassed if it had actually
    stopped, and paranoid if you'd actually jumped every time you saw a train
    coming toward your stationary train car.

    He says that even if technology create automated cars (driven by computer),
    that save 29,990 lives ... those 10 lives that are killed by computer
    failure will make the news - IE that is the problem for new technology is
    assurity of improvement.

    Actually personally though, I think its just getting the people with money
    to sell it, make it, and market it to the public, after that people will
    follow. After all if they make all new cars come with a new facility then
    people just accept it and choose whether to enable it or not. If you put out
    enough coercian via media advertising (controlling the media - various
    forms) etc, people will follow anything pretty much... they'd probably even
    launch themself into space and all get killed, if someone told us we had to
    do it cause the earth was going to self destruct and the government funded a
    'launch everyone into space' project to save us, because some scientists or
    media controllers got a misunderstanding.... like a significant size astroid
    that heads toward us for years, but changes path just at the last moment and
    actually misses us. Lucky for those that didn't know what was going on and
    stayed behind. You get the point anyway... with enough power (some kind of
    measure of the ability to control/ influence everyone), any changes can
    happen.

    Actually I think it a bit simpler than that though. Fully automated cars are
    possible - therefore they WILL happen. Just a question of when.

    JP
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-