Jump to content


  • Content count

  • Joined

  • Last visited

Community Reputation


About SilentWorld

  • Rank

Recent Profile Visitors

1,815 profile views
  1. New car!

    Very nice. I bought a Civic hatchback a couple months ago. It’s got a little turbo in it that gives it some nice pep.
  2. Isn’t it pretty much impossible to sue for injuries in this province? I guess you’d know more about that than me. Id expect the truck driver will likely face significant criminal liabilities, not sure about civil. I actually feel pretty bad for him as well, no matter where he goes somebody is going to know somebody that died in this wreck. Sure he screwed up but we’ve all done something stupid while driving it’s just usually you don’t kill 15 people because of it.
  3. My understanding is there were people trapped in the wreckage initially. I don't think the roof was torn off cleanly like that originally.
  4. But can you imagine if it did happen
  5. More likely that this is a frighteningly common occurrence in your country is the reason why it will be forgotten.
  6. Another Tesla autopilot crash

    Tesla, as it exists now, will not exist in 5 years. Anyone who’s been paying attention should have already realized this.
  7. Another Tesla autopilot crash

    This is almost certainly due to driver inattentiveness, but just like I said in the thread about the Uber crash, auto manufacturers bear some responsibility when they design systems that encourage that exact inattentiveness.
  8. Self-driving Uber vehicle kills pedestrian

    I think the primary issue here is that you really can't have an AI that is actually driving the car unless it's good enough that it can essentially always drive the car. It's unreasonable to expect a human driver to pay attention as if he were driving the car when he's not. I think that if the human driver had been paying attention, that accident likely could have been avoided (I assume the car has freaking headlights). Unfortunately, the driver was looking down the entire time. I sort of have personal experience related to this kind of automation. It's not exactly the same but we have tractors that drive in a straight line using GPS. These tractors do not make the corners at the end of the field, you have to make the corner and line it up for the next pass, but let me tell you if you're driving through a 2 mile long field at 5 MPH, you are NOT going to pay attention the whole way. You're just not. Even if you know there may be an obstacle midway through the pass that you'll have to watch out for (for example mud/standing water) you'll often start losing focus, only to suddenly snap back to attention seconds before the tractor is about to get stuck. If you were driving at 50 MPH instead of 5, and had pedestrians to watch out for instead of mud puddles that don't freaking move, that would just be asking for trouble. Automated technologies that assist the driver are good. Traction control, automated emergency braking, lane keep assist (under certain circumstances) can all be helpful if they're merely HELPING the driver control the vehicle. Systems that operate in a manner that the driver would be unable to do so (traction control applying brakes to specific wheels to increase traction for example) or taking over only when it's clear the driver has failed to notice something (emergency braking detecting an imminent collision and braking for example) are great but I think to actually create a system that can take complete control of the vehicle, but at the same time has shortfalls that require the driver to be constantly vigilant is just asking for trouble. That all being said, it is obvious that if we want to get to a point where we do have competent autonomous driving systems, that we're going to have to test systems that are not competent. That driver wasn't paying attention AT ALL, and for a car that is not yet ready to drive itself, that's ridiculous. Like I said above though, that is going to happen if you go long periods of time without having to control the vehicle. So I wouldn't put the blame entirely on the driver. Uber should have (and likely did) realize this kind of stuff was happening. Certainly, now that this death has occurred any plausible deniability is out the window. They know that their cars cannot safely operate without a driver paying attention and they also know that drivers in their automated cars will often not be paying attention at all. Now, my understanding is that there are systems that are capable of monitoring a driver's eyes and other signs in an attempt to ensure they are paying attention. I think I read that the new Cadillacs with Super Cruise have such a system. I think that such systems need to be put in place for any testing of autonomous driving. Without some sort of method for ensuring that drivers are paying attention, this kind of incident will happen again.