5 ms·

LIDAR is a crutch and a dead end. Some startups have already wisened up and are only using cameras.

by zmarty 8y ago

syntaxing 8y ago
Interesting because to me, LIDAR is an absolute necessity. Startups use cameras (+ radar) only because it's technically and financially "easier" than other technology. I don't know of any company with Level 3 and above self driving capabilities not using LIDAR (though I would love to be proved wrong because that would be awesome to see)
- ModernMech 8y ago
  In the long run, binocular cameras should be enough. We are an existence proof of that. But that requires human-level cognition and inference, which is not something that's even on the horizon. Until then, LIDAR will be a key technology.
  - raisedbyninjas 8y ago
    Even when cameras are "enough" LIDAR will still provide additional data not obtainable through cameras, allowing even safer driving. If humans had evolved biological LIDAR we would use it while driving.
    - Klathmon 8y ago
      I may be mis-remembering, but wasn't an argument from Tesla that LIDAR would actually be a disadvantage? LIDAR has a lot more "false positives" if I remember correctly, and a lot of the work is going toward filtering LIDAR data and cleaning it up in cars that use it. I could see not wanting to spend the time and effort in trying to tame something like that when they have something working pretty well right now with different kinds of sensors.
      - shaklee3 8y ago
        I think at this stage, nobody is certain. But I tend to side with more information is better, and you filter what you don't need. Same reason sensor fusion algorithms use the gyroscope and accelerometer in concert to measure movement.
        sliken 8y ago
        Dunno, seems like Lidar often results in seeing plastic bags or tumbleweeds as car killing objects.
        shaklee3 8y ago
        Yes, but that's what I was getting at. You filter those out from the camera. The reverse is probably true as well, where you need depth since an image can't give you enough information.
  - albertgoeswoof 8y ago
    Self driving cars need to be safer and faster than human driven cars, otherwise it will be cheaper to keep humans in the driving seat.
    - mrep 8y ago
      You are ignoring the cost of human labor which in the US is minimum $7.25 an hour which is why waymo is targeting taxis as they don't have to pay those human labor costs if they can get rid of the drivers.
  - xoa 8y ago
    >In the long run, binocular cameras should be enough. No they won't. They have all the issues that pure visible light sensors fundamentally have in Earth weather and lighting conditions. There hasn't been much choice for human drivers obviously (though in principle there we could have pushed for better HUD ages ago, and could still make use of AR), but with the switch to computers we can and should do better. It is foolish not to make use of better data input. >We are an existence proof of that. Proof of what? Millions of deaths and casualties every single year due to driving? We tolerate it because the benefits are even bigger, and there aren't any other options. It is literally worth mass suffering and death, not to mention economic expense, to gain arbitrary point-to-point mechanized transportation. But that acceptance is purely a relative matter because there is nothing better, as with medical technology standards can and will change once we can improve. There is no reason that self driving cars shouldn't generally be able to see animals (deer, moose etc) in a pitch black night in the fog for example and keep from running into them. Yeah that'd be impossible for us, but that's not a law of physics just a limitation of information input limited purely to the visual spectrum. Using humans as a standard is really foolish given how objectively terrible we are at this.
    - ModernMech 8y ago
      > Using humans as a standard is really foolish given how objectively terrible we are at this. Yeah, it would be foolish to use my grandmother as a benchmark for driverless cars. Humanity would experience a net increase in safety if she were to take a driverless car to church. But even the best driverless car today with all their LIDARs can't outperform my father, a 40 year veteran of UPS with literally millions of accident free miles under his belt in every driving condition imaginable - rain, sleet, snow, fog, bright sun. All with binocular cameras, and a finely tuned driving ethic. If I presented UPS with a fleet of similarly capable driverless trucks, I would be a billionaire, and the world would be net-safer.
      - mrep 8y ago
        US car fatalities are at 1.16 per 100 million miles in 2017 [0]. Anecdotal evidence for accidents based on multiple orders of magnitude less miles doesn't say much (fatalities was the first data i could find) [1]. [0]: https://en.wikipedia.org/wiki/Motor_vehicle_fatality_rate_in_U.S._by_year https://en.wikipedia.org/wiki/Motor_vehicle_fatality_rate_in... [1]: 20 mph average driving rate (guess on my part but they stop a lot) * 8 hours a day * 5 days week * 50 weeks a year * 40 years = 1.6 million miles
  - InitialLastName 8y ago
    > In the long run, binocular cameras should be enough. We are an existence proof of that. Americans kill 40,000 people a year driving with their insufficient binocular vision. It's obviously not enough.
    - ModernMech 8y ago
      Insufficient binocular vision isn't always the cause though. Not all human controllers are equally capable. There is a such thing as the best driver in the world and the worst driver in the world. For example, my father worked as a driver for UPS for 40 years. He drove literally millions of miles in every condition imaginable and never got into an accident. Not a single fender bender. My sister on the other hand has been driving for a decade and has been in dozens. They have the same sensory input but vastly different performance. I would definitely buy a driverless car that performs like my father. If I could build one like that, I would be a billionaire. Not so much with one that performs like my sister.
  - erikpukinskis 8y ago
    Human eyes take in data that's not available to a camera. Fo example we can detect a single photon. We can also change the lens shape of the eye. And the brain uses more than just eyes to drive. We use our ears for example, and our vestibular system. And there are even parts of our body at work that don't talk to the brain... Our brainstem certainly handles many aspects of driving without talking to the eyes. So I wouldn't say there's an existence proof there.
FireBeyond 8y ago
"It costs too much to be palatable".
agildehaus 8y ago
And not a one of them is anywhere near level 4 self driving.

LIDAR is a crutch and a dead end. Some startups have already wisened up and are only using cameras.

by zmarty 8y ago

LIDAR is a crutch and a dead end. Some startups have already wisened up and are only using cameras.

Interesting because to me, LIDAR is an absolute necessity. Startups use cameras (+ radar) only because it's technically and financially "easier" than other technology. I don't know of any company with Level 3 and above self driving capabilities not using LIDAR (though I would love to be proved wrong because that would be awesome to see)

ModernMech 8y ago
In the long run, binocular cameras should be enough. We are an existence proof of that. But that requires human-level cognition and inference, which is not something that's even on the horizon. Until then, LIDAR will be a key technology.
- raisedbyninjas 8y ago
  Even when cameras are "enough" LIDAR will still provide additional data not obtainable through cameras, allowing even safer driving. If humans had evolved biological LIDAR we would use it while driving.
  - Klathmon 8y ago
    I may be mis-remembering, but wasn't an argument from Tesla that LIDAR would actually be a disadvantage? LIDAR has a lot more "false positives" if I remember correctly, and a lot of the work is going toward filtering LIDAR data and cleaning it up in cars that use it. I could see not wanting to spend the time and effort in trying to tame something like that when they have something working pretty well right now with different kinds of sensors.
    - shaklee3 8y ago
      I think at this stage, nobody is certain. But I tend to side with more information is better, and you filter what you don't need. Same reason sensor fusion algorithms use the gyroscope and accelerometer in concert to measure movement.
      - sliken 8y ago
        Dunno, seems like Lidar often results in seeing plastic bags or tumbleweeds as car killing objects.
        shaklee3 8y ago
        Yes, but that's what I was getting at. You filter those out from the camera. The reverse is probably true as well, where you need depth since an image can't give you enough information.
- albertgoeswoof 8y ago
  Self driving cars need to be safer and faster than human driven cars, otherwise it will be cheaper to keep humans in the driving seat.
  - mrep 8y ago
    You are ignoring the cost of human labor which in the US is minimum $7.25 an hour which is why waymo is targeting taxis as they don't have to pay those human labor costs if they can get rid of the drivers.
- xoa 8y ago
  >In the long run, binocular cameras should be enough. No they won't. They have all the issues that pure visible light sensors fundamentally have in Earth weather and lighting conditions. There hasn't been much choice for human drivers obviously (though in principle there we could have pushed for better HUD ages ago, and could still make use of AR), but with the switch to computers we can and should do better. It is foolish not to make use of better data input. >We are an existence proof of that. Proof of what? Millions of deaths and casualties every single year due to driving? We tolerate it because the benefits are even bigger, and there aren't any other options. It is literally worth mass suffering and death, not to mention economic expense, to gain arbitrary point-to-point mechanized transportation. But that acceptance is purely a relative matter because there is nothing better, as with medical technology standards can and will change once we can improve. There is no reason that self driving cars shouldn't generally be able to see animals (deer, moose etc) in a pitch black night in the fog for example and keep from running into them. Yeah that'd be impossible for us, but that's not a law of physics just a limitation of information input limited purely to the visual spectrum. Using humans as a standard is really foolish given how objectively terrible we are at this.
  - ModernMech 8y ago
    > Using humans as a standard is really foolish given how objectively terrible we are at this. Yeah, it would be foolish to use my grandmother as a benchmark for driverless cars. Humanity would experience a net increase in safety if she were to take a driverless car to church. But even the best driverless car today with all their LIDARs can't outperform my father, a 40 year veteran of UPS with literally millions of accident free miles under his belt in every driving condition imaginable - rain, sleet, snow, fog, bright sun. All with binocular cameras, and a finely tuned driving ethic. If I presented UPS with a fleet of similarly capable driverless trucks, I would be a billionaire, and the world would be net-safer.
    - mrep 8y ago
      US car fatalities are at 1.16 per 100 million miles in 2017 [0]. Anecdotal evidence for accidents based on multiple orders of magnitude less miles doesn't say much (fatalities was the first data i could find) [1]. [0]: https://en.wikipedia.org/wiki/Motor_vehicle_fatality_rate_in_U.S._by_year https://en.wikipedia.org/wiki/Motor_vehicle_fatality_rate_in... [1]: 20 mph average driving rate (guess on my part but they stop a lot) * 8 hours a day * 5 days week * 50 weeks a year * 40 years = 1.6 million miles
- InitialLastName 8y ago
  > In the long run, binocular cameras should be enough. We are an existence proof of that. Americans kill 40,000 people a year driving with their insufficient binocular vision. It's obviously not enough.
  - ModernMech 8y ago
    Insufficient binocular vision isn't always the cause though. Not all human controllers are equally capable. There is a such thing as the best driver in the world and the worst driver in the world. For example, my father worked as a driver for UPS for 40 years. He drove literally millions of miles in every condition imaginable and never got into an accident. Not a single fender bender. My sister on the other hand has been driving for a decade and has been in dozens. They have the same sensory input but vastly different performance. I would definitely buy a driverless car that performs like my father. If I could build one like that, I would be a billionaire. Not so much with one that performs like my sister.
- erikpukinskis 8y ago
  Human eyes take in data that's not available to a camera. Fo example we can detect a single photon. We can also change the lens shape of the eye. And the brain uses more than just eyes to drive. We use our ears for example, and our vestibular system. And there are even parts of our body at work that don't talk to the brain... Our brainstem certainly handles many aspects of driving without talking to the eyes. So I wouldn't say there's an existence proof there.

FireBeyond 8y ago

"It costs too much to be palatable".

agildehaus 8y ago

And not a one of them is anywhere near level 4 self driving.