Stories
Slash Boxes
Comments

SoylentNews is people

posted by martyb on Saturday July 13 2019, @03:10AM   Printer-friendly
from the what-happens^W-crashes-in-Vegas... dept.

Back in 2017, Las Vegas' self-driving shuttle service got into a minor collision after just an hour into its year-long trial. While it truly was a minor incident and nobody got hurt, the fact that an autonomous vehicle was involved prompted the National Transportation Safety Board to launch a probe. Now, the agency has wrapped up its investigation and has revealed two probable causes for the incident. First is that the truck that collided with the shuttle didn't stop when it was supposed to, which is consistent with the local government's claim after the accident. The other is that the autonomous vehicle attendant didn't have easy access to the shuttle's manual controller.

Apparently, the truck driver thought the shuttle would stop at a "reasonable" distance from the truck. Although the shuttle did start slowing down when it was 98.4 feet away, it's not programmed to stop until it's only 9.8 feet away from obstacles. The attendant hit the emergency stop button when the vehicle was 10.2 feet away from the truck, but it clearly wasn't enough to prevent the incident.

In an interview with the investigators, the attendant said they considered switching to manual mode to move the shuttle out of the way, but they didn't have easy access to its handheld controller. [...] When the accident happened, the controller was stored in an enclosed space at one end of the passenger compartment.

https://www.engadget.com/2019/07/12/las-vegas-autonomous-shuttle-crash-probe/


Original Submission

This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
  • (Score: 1, Insightful) by Anonymous Coward on Saturday July 13 2019, @03:35AM (4 children)

    by Anonymous Coward on Saturday July 13 2019, @03:35AM (#866488)

    It is impossible to say something sarcastic about this.

    • (Score: -1, Troll) by Ethanol-fueled on Saturday July 13 2019, @03:53AM (1 child)

      by Ethanol-fueled (2792) on Saturday July 13 2019, @03:53AM (#866490) Homepage

      Mod me down, you niggers, redundant, niggers, for stating the obvious fact that early adopters will pay for their wallets -- and in the case of Boeing and self-driving cars, their lives. Niggers.

      • (Score: 2) by Freeman on Monday July 15 2019, @02:36PM

        by Freeman (732) on Monday July 15 2019, @02:36PM (#867194) Journal

        I'm pretty sure you were modded down for the unneeded flamebait surrounding your "obvious fact". Though, I'm guessing you already know that, so . . .

        --
        Joshua 1:9 "Be strong and of a good courage; be not afraid, neither be thou dismayed: for the Lord thy God is with thee"
    • (Score: 1, Touché) by Anonymous Coward on Saturday July 13 2019, @04:49AM

      by Anonymous Coward on Saturday July 13 2019, @04:49AM (#866498)

      The moral of the story: Make sure your joystick is always at hand.

    • (Score: 0) by Anonymous Coward on Saturday July 13 2019, @05:42AM

      by Anonymous Coward on Saturday July 13 2019, @05:42AM (#866505)

      The autonomous shuttle of peace strikes again!

  • (Score: 0) by Anonymous Coward on Saturday July 13 2019, @04:35AM (14 children)

    by Anonymous Coward on Saturday July 13 2019, @04:35AM (#866497)

    From the NTSB report,
    > Navya could monitor the shuttle’s performance in real time from its control center in Lyon, France.

    What is the latency for a connection between Las Vegas and Lyon, France? According to this page, https://wondernetwork.com/pings [wondernetwork.com] (I added the two cities), ping time between servers is about 157ms. But to get from the control center to the vehicle must be longer? Any ideas on a realistic time?

    If it gets too long, the claim of real time monitoring goes right out the window...

    • (Score: 0) by Anonymous Coward on Saturday July 13 2019, @06:22AM (13 children)

      by Anonymous Coward on Saturday July 13 2019, @06:22AM (#866514)

      Ask NASA the question about real time. Mars to Earth is about 8mins this time of year.

      Technically anything over 0ms, not real time. And to monitor are you thinking the whole camera feed, not just imposing objects. PING time and Amount of data is the time issue. If full camera feed - so a human outside the area can watch (like US Drones) and respond. Then a few bus would be ok, after that you start to hit the maximum flow the "pipe". Local Cellular, Los Vegas to MAE-EAST then trans-Atlantic, final loopp to France. Somewhere it will get tight.

      The better question is "What is emergency stopping window/time?" This gets us back to handbook 2-second rule. Nroaml for human is Eye-Brain-Foot-stop. For France would be Camera, transit, Eye, Brain, Hand, Button, transit, brake , Stop. Those extra steps would mostly move this out to a 2.5 to 3 sec rule.

      • (Score: 4, Insightful) by maxwell demon on Saturday July 13 2019, @07:30AM (3 children)

        by maxwell demon (1608) on Saturday July 13 2019, @07:30AM (#866523) Journal

        Technically anything over 0ms, not real time.

        No, technically, anything with a fixed guaranteed response time is a real-time system, even if that guaranteed response time is a year.

        --
        The Tao of math: The numbers you can count are not the real numbers.
        • (Score: 0) by Anonymous Coward on Saturday July 13 2019, @03:18PM (2 children)

          by Anonymous Coward on Saturday July 13 2019, @03:18PM (#866638)

          Good try, but wrong... :)

          Computing
          relating to a system in which input data is processed within milliseconds so that it is available virtually immediately as feedback, e.g., in a missile guidance or airline booking system.

          Maybe real time responces at not needed. :)

          • (Score: 0) by Anonymous Coward on Sunday July 14 2019, @10:38AM (1 child)

            by Anonymous Coward on Sunday July 14 2019, @10:38AM (#866844)

            Maybe try naming your sources next time?

            From wikipedia [wikipedia.org]:

            In computer science, real-time computing (RTC) [..] describes hardware and software systems [that] must guarantee response within specified time constraints, often referred to as "deadlines". The correctness of these types of systems depends on their temporal aspects as well as their functional aspects. Real-time responses are often understood to be in the order of milliseconds, and sometimes microseconds

            If a computer system has no set response deadlines, it cannot be called real-time, regardless of what your random dictionary says.

            • (Score: 1) by anubi on Tuesday July 16 2019, @11:23PM

              by anubi (2828) on Tuesday July 16 2019, @11:23PM (#867752) Journal

              Nothing said the response time could be measured in years, or even centuries.

              I am acting as a closed loop controller for the pH of my soil. I use sulfur to control it. The response time is in years. Isn't that still real time control, albeit at a glacial pace compared to most human activity?

              --
              "Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]
      • (Score: 0) by Anonymous Coward on Saturday July 13 2019, @08:06AM (8 children)

        by Anonymous Coward on Saturday July 13 2019, @08:06AM (#866527)

        Controlling it from France is a little bizarre. Why not do it from the same continent? Canada?

        But if the bus is going slow, the latency could be sufficient. It's just that this system sucks and was not ready for the real world.

        • (Score: 2) by RS3 on Saturday July 13 2019, @02:35PM (7 children)

          by RS3 (6367) on Saturday July 13 2019, @02:35PM (#866624)

          > It's just that this system sucks and was not ready for the real world.

          You didn't even glance at the article, did you? The truck hit the autonomous shuttle. The shuttle had fully stopped, but the truck kept going. The shuttle would have needed to back up to avoid the truck. Most humans would not do that, and if they did, they might back into someone else. The truck driver was cited.

          • (Score: 0) by Anonymous Coward on Saturday July 13 2019, @05:02PM (6 children)

            by Anonymous Coward on Saturday July 13 2019, @05:02PM (#866666)

            I think that some humans would have the sense to back up a foot or two (0.3 - 0.6 meter), which sounds like all that was needed to avoid the fender bender. An alert bus driver would realize that the truck driver had either a blind spot or wasn't looking in the direction of the bus, and didn't realize the bus had moved into the truck's path.

            Do any AVs have that much sense, to "realize" what can be seen by the drivers of surrounding vehicles? Clearly this one didn't.

            I've had a number of cases like this recently--a new water main is being installed alongside my curvy suburban street, with large excavators (on tracks), dump trucks and several smaller construction vehicles zipping back and forth. When the equipment is in the street, I'm very careful to catch the eye of the operator, to make sure he sees that I'm passing. Or, if an excavator is facing the other way, I make sure to pass quickly, so that even if the bucket of the excavator swings around, I'll be out of the way in plenty of time.

            • (Score: 3, Interesting) by RS3 on Saturday July 13 2019, @06:03PM (5 children)

              by RS3 (6367) on Saturday July 13 2019, @06:03PM (#866684)

              Yes, agreed, me too. However, some people, as seen in many youtube videos, will throw it in reverse and hit the gas without looking. Then the crash will be their fault. In this case, it was the fault of the truck driver, and I'm glad he was cited.

              (stating what seems obvious) We're living in a beta development society- machines, drugs, software, everything is being released into the public and we're the beta test site. The designers will learn from this and may include reversing for collision avoidance. I'd still like the truck driver to be cited, and he would be with video evidence (of his very bad driving).

              • (Score: 1, Interesting) by Anonymous Coward on Saturday July 13 2019, @06:40PM (4 children)

                by Anonymous Coward on Saturday July 13 2019, @06:40PM (#866689)

                > We're living in a beta development society-

                Gave you a +1. Of course I've noticed that beta testing is going on around me, but you've pointed out that this is rapidly becoming the norm. Is this an inevitable consequence of the exponential(?) growth of tech where new stuff has to be released faster and faster to beat the competition?

                A friend works at a chip foundry (he's now close to retirement) and said that Moore's law is basically the business plan. It predicts where they will have to be in a few years...and looking at the dark side, it sets the speed of the hamster wheel that the foundry staff is on.

                All this beta testing in public seems like Moore's law is turning into a dystopia. I don't like it! My state doesn't allow self-driving car testing on the road, but I'm sure there are many other new developments being tested all around me.

                • (Score: 3, Touché) by anubi on Sunday July 14 2019, @12:00AM (1 child)

                  by anubi (2828) on Sunday July 14 2019, @12:00AM (#866746) Journal

                  Once the bugs are worked out, the device is now considered "mature" and discontinued.

                  --
                  "Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]
                  • (Score: 2) by RS3 on Sunday July 14 2019, @03:29AM

                    by RS3 (6367) on Sunday July 14 2019, @03:29AM (#866784)

                    Once the bugs are worked out, the device is now considered "mature" and discontinued.

                    Too sad but too true. "Obsolete" and "Unsupported" are other words I loath.

                • (Score: 2) by RS3 on Sunday July 14 2019, @03:38AM (1 child)

                  by RS3 (6367) on Sunday July 14 2019, @03:38AM (#866785)

                  Thanks, mod-point-giving AC. :)

                  How about this explanation: as an engineer, the #1 thing I'm supposed to have on my resume is NOT how I improved a product, but how I cheapened, er, sorry, cost-minimized, something. Don't get me wrong, I'm all about efficiency in all ways, but the cost-cutting pressure never lets up.

                  At most jobs I've had the prevailing pressure from management was to push things out the door. This is making me grouchy- I'll write more another time when I'm not so very tired.

                  I'll add this: rumor has it that in the 1960s, IBM invented "vaporware" - promised things that did not yet exist, during sales negotiations and written into contracts. Not sure if that's true, but I've worked at companies that did that. I actually love a challenge, but not the pressure. I'm remembering some jovial salesmen who would sheepishly approach me after a tradeshow saying, "sorry, but I sold XYZ idea I had. Can we make that? Please please please?" His approach always worked with me, but some others not so much.

                  • (Score: 1, Interesting) by Anonymous Coward on Sunday July 14 2019, @04:33AM

                    by Anonymous Coward on Sunday July 14 2019, @04:33AM (#866800)

                    Long before IBM in the 1960s.

                    My father worked at Vought before WWII. When the Navy requested bids for new carrier based planes, Vought proposed two designs and expected that the more conventional would win the competition. Instead the "vaporware" F4U Corsair gull wing concept won (although it was proposed as eye candy)...and then they had to make it work. There were many, many problems that the engineers and flight test staff had to sort (not many known outside the company), but eventually it was made into a very good aircraft.

(1)