Why Tesla Autopilot will ultimately prove the self-driving industry leader

Discussion in 'In the News' started by evannex.com, May 24, 2018.

  1. evannex.com

    evannex.com New Member

    Joined:
    Dec 15, 2016
    Messages:
    57
    [​IMG]

    [​IMG]

    Tesla took an early lead in the race to develop vehicle autonomy, and its Autopilot system remains the state of the art. However, the technology is advancing more slowly than the company predicted - Elon Musk promised a coast-to-coast driverless demo run for 2018, and we’re still waiting. Meanwhile, competitors are hard at work on their own autonomy tech - GM’s Super Cruise, is now available on the CT6 luxury sedan.

    Is Tesla in danger of falling behind in the self-driving race? Trent Eady, writing in Medium, takes a detailed look at the company’s Autopilot technology, and argues that the California automaker will continue to set the pace.

    Every Tesla vehicle produced since October 2016 is equipped with a hardware suite designed for Full Self-Driving, including cameras, radar, ultrasonic sensors and an upgradable onboard computer. Around 150,000 of these “Hardware 2” Teslas are currently on the road, and could theoretically be upgraded to self-driving vehicles via an over-the-air software update.



    Above: In its current state, Tesla’s Autopilot requires a hands-on approach (Youtube: pqmdc9YoFVY[/MEDIA]]Tesla)

    Tesla disagrees with most of the other players in the self-driving game on the subject of Lidar, a technology that calculates distances using pulses of infrared laser light. Waymo, Uber and others seem to regard lidar as a necessary component of any self-driving system. However, Tesla’s Hardware 2 sensor suite doesn’t include it, instead relying on radar and optical cameras.

    Lidar’s strength is its high spatial precision - it can measure distances much more precisely than current camera technology can (Eady believes that better software could enable cameras to close the gap). Lidar’s weakness is that it functions poorly in bad weather. Heavy rain, snow or fog causes lidar’s laser pulses to refract and scatter. Radar works much better in challenging weather conditions.

    According to Eady, the reason that Tesla eschews lidar may be the cost: “Autonomy-grade lidar is prohibitively expensive, so it’s not possible for Tesla to include it in its production cars. As far as I’m aware, no affordable autonomy-grade lidar product has yet been announced. It looks like that is still years away.”

    If Elon Musk and his autonomy team are convinced that lidar isn’t necessary, why does everyone else seem so sure that it is? “Lidar has accrued an aura of magic in the popular imagination,” opines Mr. Eady. “It is easier to swallow the new and hard-to-believe idea of self-driving cars if you tell the story that they are largely enabled by a cool, futuristic laser technology…It is harder to swallow the idea that if you plug some regular ol’ cameras into a bunch of deep neural networks, somehow that makes a car capable of driving itself through complicated city streets.”

    Those deep neural networks are the real reason that Eady believes Tesla will stay ahead of its competitors in the autonomy field. The flood of data that Tesla is gathering through the sensors of the 150,000 or so existing Hardware 2 vehicles “offers a scale of real-world testing and training that is new in the history of computer science.”

    Competitor Waymo has a computer simulation that contains 25,000 virtual cars, and generates data from 8 million miles of simulated driving per day. Tesla’s real-world data is of course vastly more valuable than any simulation data could ever be, and the company uses it to feed deep neural networks, allowing it to continuously improve Autopilot’s capabilities.

    A deep neural network is a type of computing system that’s loosely based on the way the human brain is organized (sounds like the kind of AI that Elon Musk is worried about, but we’ll have to trust that Tesla has this under control). Deep neural networks are good at modeling complex non-linear relationships. The more data that’s available to train the network, the better its performance will be.

    “Deep neural networks started to gain popularity in 2012, after a deep neural network won the ImageNet Challenge, a computer vision contest focused on image classification,” Eady explains. “For the first time in 2015, a deep neural network slightly outperformed the human benchmark for the ImageNet Challenge…The fact that computers can outperform humans on even some visual tasks is exciting for anyone who wants computers to do things better than humans can. Things like driving.”

    By the way, who was the human benchmark who was bested by a machine in the ImageNet Challenge? Andrej Karpathy, who is now Director of AI at Tesla.

    ===

    Note: Article originally published on evannex.com by Charles Morris; Source: Medium

    Article: Why Tesla Autopilot will ultimately prove the self-driving industry leader
     
    • Informative Informative x 1
  2. Nothing in the article shows that a system without LIDAR is as good or better than a system with LIDAR. Just cheaper. In developing FSD cheaper can be deadlier. WAYMO and GM will probably beat Tesla. And to correct your article, Tesla was originally going to do the Cross country try by the end of 2017. Musk originally said that he dropped MobilEye because they were taking too long, but the hardware/software change has probably increased the time to EAP to years and FSD to decades.
     
  3. bobhatt2000

    bobhatt2000 Guest

    The article also dismisses simulated cars/miles are not as valuable as tesla's real world miles. You can put money that tesla is also using/used simulated data to test and improve their self driving neural networks. If I find out they have not used simulation data (and real world data) to help train their self driving system I will not use auto pilot. When I eventually get a Tesla.
     
  4. Teslaliving

    Teslaliving Moderator

    Joined:
    May 20, 2016
    Messages:
    310
    Location:
    Massachusetts, USA
    Home Page:
    I drove the latest S75D the other day and I was stunned at how bad it was in the right lane on a 3 lane highway. Almost every exit it would swing right and then go back to the left of the lane again. Seems pretty simple to know the line on the left is the one to follow on a divided highway in the US. Middle and left lanes were fine but the right is not good at all with the latest AP.
     
  5. Bone

    Bone New Member

    Joined:
    Dec 16, 2016
    Messages:
    4
    I also believe in a future of self driving without LIDAR. The only reason why it doesn’t work today is the software and probably the computing power.

    If the software and hardware would be able to interpret the information coming through the cameras alone, the car would be perfectly capable of self driving.

    Otherwise we humans could not drive!

    We only have eyes and are capable of judging distance and speed difference just by interpreting what we see!

    Why shouldn‘t a computer be able to do the same?

    By the way, I also believe that a truly autonomous car has to be able to drive without a permanent online connection or precise maps since it can not be ensured that this data is available.
     
  6. Bone

    Bone New Member

    Joined:
    Dec 16, 2016
    Messages:
    4

    Since four years I am driving a Tesla model S, actually my second one since 2017. I also find it a bit frustrating how slow the progress on AP2 is happening.

    Still I have hope that we will see massive improvement within the next two years.

    Actually the slow progress suggests that Tesla is on the right track.

    If you would explicitly hard Code the rules in the software you would have quicker progress in the beginning (Waymo, Uber, etc...) but come to a state, where you will not be able to exceed 90% meaning that your effort to cover all the needed decisions making would be infinite.

    With the neural network and deep learning you see almost no progress for a long time. Then though, it takes off on a e-curve, learning and adapting all the time getting quickly better and better.

    At Tesla this approach is also heavily supported by huge amounts of data coming from vehicles operated globally where the net can learn from the Tesla drivers.

    The others take test drives in sub urban areas where you basically have no traffic
     
  7. linneau82975

    linneau82975 Guest

    I don\'t think autopilot will ever be fail safe. It may in fact reduce accidents but there will still be unforced accidents and deaths due to the technology..
     

Share This Page