Notices
Results 1 to 12 of 12

Thread: Cars that drive themselves.

  1. #1 Cars that drive themselves. 
    Forum Freshman
    Join Date
    Apr 2009
    Posts
    5
    In the Science Channel Documentary 2057, they talk about cars that drive themselves. They say that it's a good fifty years away, but I find that to be hard to believe, mostly because I've seen (and even operated on a small scale) this technology first-hand.

    It's called an AGV (Automated Guided Vehicle). When I was in high school, I took a class called Technology Discovery, and one unit was on the AGV. Its course and work schedule are literally as easy to edit as the contents of a spreadhseet. If it finds a barrier, it will take a detour. Also, you can program it to stay on a dark black line (a "road" of sorts). Also, it has two methods of traffic control: Remote sensing via sonic waves, infrared waves, and even bumpers (as a last resort to try and stop the machines), which means the chances of you having a car crash are slim. Normally, if the other two sensing systems fail, you'll play a nice round of bumper cars with the other guy. The only problem is that this technology is so expensive that only companies can use it.

    For the other half of the equation, we have GPS navigation technology. This basically tells you what route to go to get to your destination, but you are responsible for driving there. If you make a wrong turn, the GPS navigation system will recalculate your course in a matter of seconds. Granted, they are still working out all the kinks, but it's nowhere near the price of an AGV, so I guess you could say that you're paying for less bugs when it comes to an AGV. Besides, the Tom Tom brand of GPS is actually quite reliable as far as staying up-to-date is concerned, so I've heard.

    Did it ever occur to anyone that we could just combine these two technologies, and make a smart car? The GPS technology automatically programs the AGV car, and the AGV car goes where it is told, and informs the GPS of any obstacles it encounters, so the GPS can rework the navigation, and the cycle continues.

    I honestly don't see why guys like Donald Trump and Bill Gates (ESPECIALLY Gates, since this is a matter of COMPUTERS) haven't already gotten these smart cars, because for the life of me, aside from the cost, I cannot figure out a problem with it.


    Reply With Quote  
     

  2.  
     

  3. #2  
    Suspended
    Join Date
    Apr 2007
    Location
    Pennsylvania
    Posts
    8,822
    I think that what you will discover is that there is a huge difference between having a concept that sorta works to a mature technology that you are willing to stake your life on.

    An analogy could be made to nuclear power. In the early days of nuclear power, people thought that it would make electricity "too cheap to meter." People thought we would be flying around in nuclear airplanes.


    Reply With Quote  
     

  4. #3  
    Forum Freshman
    Join Date
    Apr 2009
    Posts
    5
    Quote Originally Posted by Harold14370
    I think that what you will discover is that there is a huge difference between having a concept that sorta works to a mature technology that you are willing to stake your life on.

    An analogy could be made to nuclear power. In the early days of nuclear power, people thought that it would make electricity "too cheap to meter." People thought we would be flying around in nuclear airplanes.
    Okay, care to elaborate? What's the flaws in my argument?
    Reply With Quote  
     

  5. #4  
    Forum Masters Degree Numsgil's Avatar
    Join Date
    Jan 2009
    Posts
    708
    There's a reason teenagers are allowed to drive and children aren't. You can demonstrate that an 11 year old can drive a car just fine. It's when things go horribly, horribly wrong that you want an adult in the driver's seat. Likewise getting a computer to drive a car around isn't hard. It's when something goes horribly, horribly wrong that things get trickier.

    Humans are good at this sort of thing in a way that computers just aren't. But while it's been hard to put a computer into a real world situation, it hasn't been as hard to put telepresence in humans. I foresee a day where, like fastfood drive-through order takers and telemarketers, we farm out driving to places like India through telepresence. You'd get in your car to go to work, and dial into a rent-a-driver program, and someone in Mumbai would drive you to work. Maybe it would even be possible with a little computer assistance for one driver to drive multiple vehicles.
    Reply With Quote  
     

  6. #5  
    Forum Freshman
    Join Date
    Apr 2009
    Posts
    5
    Quote Originally Posted by Numsgil
    There's a reason teenagers are allowed to drive and children aren't. You can demonstrate that an 11 year old can drive a car just fine. It's when things go horribly, horribly wrong that you want an adult in the driver's seat. Likewise getting a computer to drive a car around isn't hard. It's when something goes horribly, horribly wrong that things get trickier.
    And who said there won't be humans nearby? Last I checked, the only purpose of a car was to transport humans from point a to point b. If something goes wrong, the "passenger" that the smart car is carrying will be on hand to troubleshoot.
    Reply With Quote  
     

  7. #6  
    Forum Masters Degree Numsgil's Avatar
    Join Date
    Jan 2009
    Posts
    708
    If you're talking commercial application then I think automated trunks transporting freight is the real application. That means transporting thing which do not know how to drive, like gasoline or frozen carrots.

    But even that aside, if your car is driving you around, you're not going to be paying attention. In an emergency, there's almost always only a few split seconds to react. If you're distracted because you're reading the newspaper while your car drives you to work, you're not going to be alert enough to react in the .2 second window you have.
    Reply With Quote  
     

  8. #7  
    Suspended
    Join Date
    Apr 2007
    Location
    Pennsylvania
    Posts
    8,822
    Quote Originally Posted by computergeek3000
    Okay, care to elaborate? What's the flaws in my argument?
    All right. For starters the GPS system isn't nearly accurate enough to drive a car. If you own a GPS, and have used it for tasks like geocaching, you know that they are only accurate to plus or minus about 30 feet at best, sometimes 50 feet if your satellite signals are weak. You will need something that controls the vehicle to maybe plus or minus 5 feet to keep it on the road.

    Now you have to deal with obstructions in the road. Maybe some kind of radar detection system. There are such things available, but is the resolution good enough to replace a human driver? Could it pick up a deer that is headed for your path in the road, but not quite there yet, and take correct evasive action? If it doesn't, somebody could get killed.

    It would have to be super reliable. If a power supply or circuit card failed, or there were some bug in the software, somebody could get killed. I think you are going to need something orders of magnitude better than what's available today.
    Reply With Quote  
     

  9. #8  
    Forum Freshman
    Join Date
    Apr 2009
    Posts
    5
    Quote Originally Posted by Harold14370
    It would have to be super reliable. If a power supply or circuit card failed, or there were some bug in the software, somebody could get killed. I think you are going to need something orders of magnitude better than what's available today.
    Two words:

    strict liability
    Reply With Quote  
     

  10. #9  
    Forum Masters Degree Numsgil's Avatar
    Join Date
    Jan 2009
    Posts
    708
    One word: Therac-25.
    Reply With Quote  
     

  11. #10  
    Forum Freshman
    Join Date
    Apr 2009
    Posts
    5
    Quote Originally Posted by Numsgil
    One word: Therac-25.
    That article never said anything about how the manufcaturers of that machine got off the hook from strict liability against those three estates.

    When I mentioned strict liability, I meant to communicate to you the fact that any company that releases smart car technology to the public will understand and accept the consequences of any malfunctions in their technology.

    Computer software right now is so unreliable because Microsoft releases these things in Beta forms (Windows Vista is a prime example), knowing that no physical injury can come from their software, and with the mindset that they could always patch it up later. When people's lives are literally at stake, Microsoft will be drastically more careful in their development. They'll even have small-scale tasts to make sure it works perfectly.

    Also, no one ever said the transition from 100% human-powered to 100% computer-powered would be instantaneous. That same documentary that I cited in my OP made several mentions of the cars realising when the driver is tired, stressed, or anything like that, and overriding the car. This means that the human and the car will be working in cohesion. You won't be able to just sit back and read a book for another hundred years at the VERY least.

    If anything else, how about this: Smart PLANES! Sky traffic isn't nearly as clogged as rush-hour ground traffic, and birds crashing into the planes has already been delt with with strong shielding.
    Reply With Quote  
     

  12. #11  
    Forum Masters Degree Numsgil's Avatar
    Join Date
    Jan 2009
    Posts
    708
    My point with Therac-25 is that bad software can kill people. And yet there's still a place for software in life critical applications. It just means that that software has to be very very very well tested, and very very very robust to weird input. Software meant for desktop computers and software meant for Martian rovers have very different testing procedures.

    The real world of traffic, however, is simply way more complex than any computer can handle with reasonable safety. For present state of the art, check out this article. That's a lot of failure in a barren desert.

    I do think there'll be a slow transition towards automated driving, though, along a gradient. It started with auomatic transmission and cruise control. Right now in San Diego there's an ongoing experiment with cars driving on the freeway in a special "auomated lane". The computers have faster reflexes, so the cars have less following distance, so you can put more cars in the lane. Likewise they network so merging/unmerging is potentially less likely to cause traffic jams. But it's still a bit far away from production ready. I won't hazard a guess for when, but eventually there should be cars that can drive themselves just fine. Someday.

    But in the immediate future I think telepresence is a viable solution. A computer program, especially one with lives at stake, is expensive to produce. Compared with relatively plentiful and cheap human labor in places like India. It's not as sexy, but it amounts to about the same thing. Sort of like a virtual taxi cab driver. You get in your car, pay a fare, and someone in Bangladesh drives you wherever you need to go.
    Reply With Quote  
     

  13. #12  
    Suspended
    Join Date
    Jan 2008
    Posts
    440
    Radar...

    Every car would need to have a system using it's own frequency, it would need to be 'unjammable' it would need to resolve down to a few millimeters, it would need to resolve texture, it would need to resolve material, temperature, motion.

    You would not swerve to avoid what you think is a paper bag but you might for a brick, thus your system would need to be able to tell the difference.

    Today no such system exists, the best the very best military radar can tell the difference between an 8inch diameter gun muzzle and and 8inch diameter sewer pipe at an unpublished range, probably a kilometre or so, and costs rather a lot, weighs around 100Kilos and needs the very fastest computers to process the data

    That is not going to happen, the potential to 'jam' such a system is huge, just obtain a similar radar and tweak it and bingo you have blinded them.

    Public transport on rails is about all we have the technology for today. But even the latest, fastet trains whether maglev or other have a driver.

    The biggest barrier is that you will need to boot all your computers with an operating system that does not crash (pun intended), would you trust your life to what we have today?

    Somehow,

    "Warning this computer has encountered a fatal error and will close immediately"

    Whilst travelling at 130MPH in the rain between two 18 wheelers with the doors and windows controlled by the PC is not a message I'd like to see


    As for 'strict liablilty' companies need only show they had not cut corners, that all precautions had been taken, all guidlines followed, all systems checked etc etc and all their work was traceable, that they had taken "every reasonable care" to ensure their product was safe.

    IF you prosecuted every design engineer who ever made a mistake technology would cease to improve beyond what it is today.
    Reply With Quote  
     

Bookmarks
Bookmarks
Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •