7 ms·

GM is already ahead of Tesla in vehicle autonomy. It's even debatable as to whether GM is ahead of Waymo. Waymo is poised to deploy first in Phoenix, but they h

GM is already ahead of Tesla in vehicle autonomy. It's even debatable as to whether GM is ahead of Waymo. Waymo is poised to deploy first in Phoenix, but they haven't even begun serious testing in dense urban areas, which are far more challenging and where Cruise is focused and making headway. GM may or may not achieve it's targets on time, but without question they are putting their money where there mouth is. GM is the most vertically integrated and they're showing very good progress relative to the competition, they're in the best position to deploy at scale. Their acquisition of Cruise was money well spent. GM has a legit opportunity to achieve dominance. And really, if anybody is spreading hype and BS about autonomous vehicles it's Tesla. Developing an autonomous OS to minimum viable product is a ~5 year job, depending on how well or poorly things go. Musk was tweeting in the spring that they'd be able to demonstrate L5 capabilities on underpowered hardware by the end of this year. Which is total bullshit. Many car companies are making nebulous claims about when they expect to achieve autonomous capabilities, but Tesla lied outright to get people to reserve their Model 3 orders.

tonyhb 9y ago
> Waymo is poised to deploy first in Phoenix, but they haven't even begun serious testing in dense urban areas I've seen Waymo cars in SF around Market st, near Dolores Park and near Noe Valley. I assume that they're also testing heavily in their main cities, so (anecdotally) I don't think this is accurate.
- maxerickson 9y ago
  We'll find out in a few months when the 2017 California disengagement reports come out. In 2016, Waymo had ~630,000 autonomous miles and GM Cruise had a few thousand, so there isn't a comparison to be made on that data. https://www.dmv.ca.gov/portal/dmv/detail/vr/autonomous+/testing https://www.dmv.ca.gov/portal/dmv/detail/vr/autonomous+/test...
- Fricken 9y ago
  Waymo has been in and out of San Francisco since the beginning, but as far as I know it's not full-scale testing in preparation for a deployment. Waymo either is, or has been in 14 different cities, but their primary focus is in Phoenix.
  - bduerst 9y ago
    >their primary focus is in Phoenix. For the level 4 driverless tests, yes. GM isn't even close to L4 driverless testing yet. Also, Waymo has gathered enough road data that they're doing accelerated testing in simulation now. I think you might be underestimating the complexity of the self driving car tech when doing your comparison.
    - Fricken 9y ago
      The subject here isn't GM's in-house autonomous development program (Supercruise), it's their subsidiary Cruise Automation, who has been doing L4 testing for at least 18 months now. Cruise Automation began in 2013 as a silicon valley robotics start-up, and was acquired by GM in early 2016. Cruise is also using simulation. They mentioned in their presentation yesterday that they have 3 simulation engines, and together they do 150 simulations/minute. I'm not sure exactly what that means but simulation is an integral part of their validation process just like with Waymo.
      - bduerst 9y ago
        My point is that your comparing progress by measuring the number of cities they're in is senseless, because at a certain point road data stops and simulations take over. Also, GM nor Cruise have launched driverless L4 testing yet, which Waymo was the first company (ever) to do last month: https://arstechnica.com/cars/2017/11/fully-driverless-cars-are-here/ https://arstechnica.com/cars/2017/11/fully-driverless-cars-a...
        danblick 9y ago
        My understanding is that "driverless testing" here means either (1) moving the safety driver to the back seat or (2) following the vehicle with a second vehicle with a remote kill switch. Thanks for posting that link though, I've been curious to read more about this. """At first, most of Waymo's driverless cars will have an employee in the back observing the vehicle's behavior. If something goes really wrong, they'll be able to push the "pull over" button to stop the car."""
bmuon 9y ago
I'm seriously concerned about this. "Self-driving" is not a library that you get from Github and just dump it into your car. Waymo has been doing this for a decade. Do GM's claims mean that Waymo's experience doesn't matter much or is GM actually going to cause real world accidents by moving too fast?
- maxerickson 9y ago
  They're serious. https://medium.com/kylevogt/why-testing-self-driving-cars-in-sf-is-challenging-but-necessary-77dbe8345927 https://medium.com/kylevogt/why-testing-self-driving-cars-in...

tonyhb 9y ago

> Waymo is poised to deploy first in Phoenix, but they haven't even begun serious testing in dense urban areas I've seen Waymo cars in SF around Market st, near Dolores Park and near Noe Valley. I assume that they're also testing heavily in their main cities, so (anecdotally) I don't think this is accurate.

maxerickson 9y ago
We'll find out in a few months when the 2017 California disengagement reports come out. In 2016, Waymo had ~630,000 autonomous miles and GM Cruise had a few thousand, so there isn't a comparison to be made on that data. https://www.dmv.ca.gov/portal/dmv/detail/vr/autonomous+/testing https://www.dmv.ca.gov/portal/dmv/detail/vr/autonomous+/test...
Fricken 9y ago
Waymo has been in and out of San Francisco since the beginning, but as far as I know it's not full-scale testing in preparation for a deployment. Waymo either is, or has been in 14 different cities, but their primary focus is in Phoenix.
- bduerst 9y ago
  >their primary focus is in Phoenix. For the level 4 driverless tests, yes. GM isn't even close to L4 driverless testing yet. Also, Waymo has gathered enough road data that they're doing accelerated testing in simulation now. I think you might be underestimating the complexity of the self driving car tech when doing your comparison.
  - Fricken 9y ago
    The subject here isn't GM's in-house autonomous development program (Supercruise), it's their subsidiary Cruise Automation, who has been doing L4 testing for at least 18 months now. Cruise Automation began in 2013 as a silicon valley robotics start-up, and was acquired by GM in early 2016. Cruise is also using simulation. They mentioned in their presentation yesterday that they have 3 simulation engines, and together they do 150 simulations/minute. I'm not sure exactly what that means but simulation is an integral part of their validation process just like with Waymo.
    - bduerst 9y ago
      My point is that your comparing progress by measuring the number of cities they're in is senseless, because at a certain point road data stops and simulations take over. Also, GM nor Cruise have launched driverless L4 testing yet, which Waymo was the first company (ever) to do last month: https://arstechnica.com/cars/2017/11/fully-driverless-cars-are-here/ https://arstechnica.com/cars/2017/11/fully-driverless-cars-a...
      - danblick 9y ago
        My understanding is that "driverless testing" here means either (1) moving the safety driver to the back seat or (2) following the vehicle with a second vehicle with a remote kill switch. Thanks for posting that link though, I've been curious to read more about this. """At first, most of Waymo's driverless cars will have an employee in the back observing the vehicle's behavior. If something goes really wrong, they'll be able to push the "pull over" button to stop the car."""

bmuon 9y ago

I'm seriously concerned about this. "Self-driving" is not a library that you get from Github and just dump it into your car. Waymo has been doing this for a decade. Do GM's claims mean that Waymo's experience doesn't matter much or is GM actually going to cause real world accidents by moving too fast?

maxerickson 9y ago
They're serious. https://medium.com/kylevogt/why-testing-self-driving-cars-in-sf-is-challenging-but-necessary-77dbe8345927 https://medium.com/kylevogt/why-testing-self-driving-cars-in...