Stories
Slash Boxes
Comments

SoylentNews is people

posted by Fnord666 on Friday December 01 2017, @02:48AM   Printer-friendly
from the need-to-drive-waymo-miles dept.

Waymo racks up 4 million self-driven miles

Waymo continues to press its lead in terms of actual miles driven on roads, which is potentially the most important metric out there when it comes to building successful autonomous driving technology. The Alphabet-owned company that began life as Google's self-driving car project around a decade ago now has 4 million miles driven autonomously on roads.

That 4 million miles represents the self-driving effort of Waymo's entire test fleet, covering its original autonomous vehicles all the way up to its current driverless Chrysler Pacifica minivans, which are actually now testing on Arizona public roads, right alongside everyday human drivers, with no safety driver behind the wheel at all.

In simulations, Waymo's bots have driven 2.5 billion "virtual miles".

Also at The Verge.


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This discussion has been archived.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.
  • (Score: 5, Insightful) by Runaway1956 on Friday December 01 2017, @03:42AM (6 children)

    by Runaway1956 (2926) Subscriber Badge on Friday December 01 2017, @03:42AM (#603786) Journal

    Early on, people were telling us how safe autonomous cars would be - but they had no statistics to back them up. 4 million miles isn't a lot - actually, it isn't even much - but they BEGIN to mean something. Around 20 million miles, we can actually start taking the numbers seriously. How many accidents, how many deaths, how many injuries, how much property damage, etc, etc - just like the insuarance companies measure people's driving. There is sure to be a difference, but what differences, and how significant those differences are remains to be seen.

    More importantly, how quickly do the engineers and developers react to new data and mishaps? I remember the story of one of those semi-autonomous cars almost backing over a pedestrian while parking. The explanation that the car didn't have pedestrian sensing features was pretty damned lame. If they advertised the vehicle as self-parking, then you EXPECT the car to be able to park itself, without running over some nitwit standing in the way.

    Real life conditions in the real world, with many millions of miles of real life driving is what will sell autonomous cars. Test conditions on sheltered tracks mean almost nothing.

    Starting Score:    1  point
    Moderation   +3  
       Insightful=3, Total=3
    Extra 'Insightful' Modifier   0  
    Karma-Bonus Modifier   +1  

    Total Score:   5  
  • (Score: 1, Touché) by Anonymous Coward on Friday December 01 2017, @05:47AM (1 child)

    by Anonymous Coward on Friday December 01 2017, @05:47AM (#603816)

    If they advertised the vehicle as self-parking, then you EXPECT the car to be able to park itself, without running over some nitwit standing in the way.

    But it begs the question: wouldn't the world be better with fewer nitwits?

    • (Score: 2) by bob_super on Friday December 01 2017, @08:07PM

      by bob_super (1357) on Friday December 01 2017, @08:07PM (#604022)

      Actually, the problem is: will teh car park despite the chair or bucket put there by the person who shoveled the space?
      Until that code is reliable, letting your car self-park in winter in many neighborhoods would result in highly unpleasant results...

  • (Score: 1) by galgon on Friday December 01 2017, @01:32PM (2 children)

    by galgon (3041) on Friday December 01 2017, @01:32PM (#603885)

    I want self driving cars everywhere in a few years. With that kind of goal 1 Million miles in 6 months is too slow. They need to move faster. Waymo has 100 vans driving around Phoenix. Each van should be doing at LEAST 200 miles a day. Which puts you at 1 million miles every 2 months. Waymo is being very conservative in their rollout process. If Uber or Tesla were in the same position we would have 1000s of cars on the road by now.

    • (Score: 2) by Runaway1956 on Friday December 01 2017, @02:41PM (1 child)

      by Runaway1956 (2926) Subscriber Badge on Friday December 01 2017, @02:41PM (#603901) Journal

      200 miles per day - in a city? That's a lot of driving. And, probably pointless driving. You do realize that a lot of city dwellers may not drive 200 miles in a month? A hundred miles in city traffic is a much longer drive than someone like me, who can decide that he wants to go to Dallas, or Tulsa, and just run down the interstate, or going north, hop on a primary highway for several miles, then get on the turnpike the rest of the way.

      If they're running autonomous vehicles in Phoenix, I certainly HOPE there is some point to the running. Carry people to work, school, doctor's office, shopping, whatever, not just drive around for the sake of racking up miles.

      • (Score: 1) by galgon on Saturday December 02 2017, @04:12AM

        by galgon (3041) on Saturday December 02 2017, @04:12AM (#604139)

        NYC cabs run about 200 miles a day. Phoenix is a larger city and likely more miles per trip on average. If they were using these like an Uber they could easily do 200 miles a day. Unfortunately they are still trialing them with a small group of test families. Hopefully they open it up to the public soon.

  • (Score: 3, Insightful) by DannyB on Friday December 01 2017, @02:36PM

    by DannyB (5839) Subscriber Badge on Friday December 01 2017, @02:36PM (#603896) Journal

    Miles Per Accident would be a useful statistic to publish.

    The sheeple are going to also need some measure of battery capacity that makes more sense than killowat-hours. That "hours" part is going to confusicate people. Just like 'light-year' isn't a unit of time.

    --
    To transfer files: right-click on file, pick Copy. Unplug mouse, plug mouse into other computer. Right-click, paste.