Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Waymo CEO dismisses Tesla self-driving plan: “This is not how it works”

arstechnica.com

Post reply on HN
Print view View original

511–520 of 581 posts

Re: Waymo CEO dismisses Tesla self-driving plan: “This is not how it works”

#511

Post by mongol »

crazygringo wrote: post #500
mongol wrote: post #305

Earlier quoted context omitted.

I also don't use cruise control. Not because I don't trust it, but I don't trust myself to react as quickly with it on. If my "gas foot" rests, it will take longer for it to hit the brake if needed, than if I engage continuously while driving. No scientific proof for this, except my own perception of my own attention.

That's really interesting. Funny, I don't feel the same -- my foot rests below the brake (on the floor) instead of on the gas pedal so the time to move it to the brake feels roughly the same, and I'm paying 100% attention while driving since I'm still steering and constantly monitoring distance to the cars in front of me. So I don't notice any less attention -- it's just a rest for the muscles in my right foot. But r…

When I have tried to find a place for my foot when driving cruise control, it has happened that I got the toes of my foot stuck under the brake pedal. It was like I was resting the foot partly below it so when I moved it up, it was in the way. Can differ between cars surely. But in the cars I have driven with cruise control, I have not found a comfortable and at the same time "alert" location to put my foot.

Re: Waymo CEO dismisses Tesla self-driving plan: “This is not how it works”

#512

Post by screye »

_ea1k wrote: post #163
screye wrote: post #35

I worked with a self-driving product team a couple of years ago. Tesla's approach was looked upon with suspicion and often openly mocked. Tesla's entire self-driving strategy is led by Karpathy, who while brilliant, is an under-experienced researcher with a narrow expertise in large scale 2D CNNs. Tesla (and Comma.ai ?) seem to be the only groups that find Lidar unnecessary for L5 self-driving. I completely agree wit…

I'm pretty sure mobileye has demonstrated their FSD solution without lidar as well. It seems like they are treating lidar as optional. They have many more cameras than Tesla to support the camera only approach, though.

At some point, you have enough cameras that you are probably (implicitly or explicitly) doing 3D reconstructions that Lidar does anyways. At that point, why not just use LIDAR ?

To save some $2-3k? That feels 'penny wise pound foolish'. This is a premium feature and LIDAR keeps getting cheaper.

Re: Waymo CEO dismisses Tesla self-driving plan: “This is not how it works”

#513

Post by leesec »

"For us, Tesla is not a competitor at all," Krafcik said. "We manufacture a completely autonomous driving system. Tesla is an automaker that is developing a really good driver assistance system."

This is absolutely hilarious to me. What an amazing cope. To downplay Tesla's approach is something you'd expect from a competing CEO. To straight up declare they are not even a competitor is delusional. They are clearly competitors.

Re: Waymo CEO dismisses Tesla self-driving plan: “This is not how it works”

#514

Post by BingoAhoy »

necovek wrote: post #412
roenxi wrote: post #370

Earlier quoted context omitted.

There is no challenge reconciling imperfect FSD with high trust of that FSD. When I decided to play the game of risk minimisation, I sold my car. Minimising risk isn't the most important goal of drivers, almost by definition. Cars are not safe in any objective sense. They are tools of convenience. A fun hypothetical, you and a good friend get tested for spatial intelligence and it turns out there was a big difference…

While spatial awareness is important during driving,I believe being focused on driving is even more so. When driving the tight streets of old European cities with pedestrians jumping out everywhere, I usually watch for hints like too tall cars parked on the sidewalk and potentially hiding pedestrians planning to cross the street, and move my foot from the gas to hovering above the brake pedal. And a million other thi…

You mention focused driving, but here's a cool idea. Your subconscious which actually handles most of ur behavior and decision making and nuanced calculations gradually learns from your conscious. When you focus on things, you gradually train your subconscious to mirror that behavior and do it autonomously.

This is demonstrable by reflecting on new things you learn versus old things. Old things like walking you barely put any conscious effort in, yet once you reach a certain age the daily obstacle course that is life, which is full of tripping hazards becomes effortless to avoid ensnaring ur foot on and succumbing to a sudden tumble. But if you were to try to roller blade for the first time, suddenly you have to put massive conscious strain and focus on every movement just to avoid falling on something so simple as a slight texture change on a surface.

Also interesting thought on (conscious) spatial awareness: Here's a question is your conscious aware of things first or is your subconscious aware first? When you conscious becomes aware how sure are yah that it's not your subconscious first alerting your conscious beforehand? These are rhetorical questions which psychologists and neuroscientists already have insights about :).

Life is dangerous, but many of the dangers are predictable and the brain is adept growing to adjust to that predictability AND at learning to recognize indicators for unpredictable dangers(humans receive anxiety in these moments). In those latter situations Intelligence and consciousness is needed. Dangers that are predictable can be learned to be subconsciously handled without much worry & with much practice + experience.

Tesla autopilot is a computerized subconscious that's consciously trained by all the tesla drivers.

I strongly suspect that we'll never have level 5 autopilot with or without lidar sensors unless the computers get a human adaptable intelligence module OR some convention simplifies the environment such that new unpredictable dangers can be minimized to a miniscule and acceptable failure rate. I think people in this debate are focusing on the wrong issues.

Re: Waymo CEO dismisses Tesla self-driving plan: “This is not how it works”

#515

Post by sudosysgen »

Erlich_Bachman wrote: post #458
sudosysgen wrote: post #356

Earlier quoted context omitted.

The human eye has capabilites that you'd need a camera more expensive than a LiDAR by a good bit too replicate. No rolling shutter, very low delay, continuous signal processing instead of per-frame, maximum resolution around ~70MP eq. in the center, f/2.8 aperture at a full-frame image size, servo driven active rangefinding, continuous cleaning, etc... A camera with those features would be around 3000-15000$, and you…

> Also, the human brain can use focus information and stereo phase to deduce 3D structur Current AI tech can do this too, from multiple images. Karpathy even talked about how they do use that in Teslas.

The human eye does it in a very different way. It combines both stereo, parallax, eye convergence, and focus.

Tesla only does the first.

Re: Waymo CEO dismisses Tesla self-driving plan: “This is not how it works”

#516

Post by tim333 »

justapassenger wrote: post #62
screye wrote: post #35

I worked with a self-driving product team a couple of years ago. Tesla's approach was looked upon with suspicion and often openly mocked. Tesla's entire self-driving strategy is led by Karpathy, who while brilliant, is an under-experienced researcher with a narrow expertise in large scale 2D CNNs. Tesla (and Comma.ai ?) seem to be the only groups that find Lidar unnecessary for L5 self-driving. I completely agree wit…

You can see in all public talks how Karpathy focuses almost exclusively on image recognition. Yeah, that’s great and important, but that’s a step one in making any autonomous vehicle. And the easiest one.

>And the easiest one

Some of Telas biggest problems seem to stem from not recognising stationary trucks in their path.

Re: Waymo CEO dismisses Tesla self-driving plan: “This is not how it works”

#517

Post by sudosysgen »

jfoster wrote: post #398
sudosysgen wrote: post #356

Earlier quoted context omitted.

The human eye has capabilites that you'd need a camera more expensive than a LiDAR by a good bit too replicate. No rolling shutter, very low delay, continuous signal processing instead of per-frame, maximum resolution around ~70MP eq. in the center, f/2.8 aperture at a full-frame image size, servo driven active rangefinding, continuous cleaning, etc... A camera with those features would be around 3000-15000$, and you…

You don't need human eye equivalents. Human eyes are probably not the ideal cameras for driving.

This may be the case, you might not have to reproduce everything.

But just to give an example that is very relevant to self-driving, getting a camera with similar low-light video performance as the human eye costs over 2000$.

Re: Waymo CEO dismisses Tesla self-driving plan: “This is not how it works”

#518

Post by Spooky23 »

pas wrote: post #382
Spooky23 wrote: post #193

Earlier quoted context omitted.

Frankly, it always sounded like a solution looking for a problem for me. Transportation is like electricity, it’s all about peak demand. I’m writing this at midnight on a Sunday... there’s no demand for transportation. Come 7-930AM, demand is 1000x more. There’s no self driving taxi that makes sense as car replacement. We already have busses, which are already really cheap and can get cheaper if cars are too expensiv…

Current peak demand in cars is limited by roads, and ... always will be. (Hence Musk trying to dig tunnels for more roads.) Which might be okay if throughput is kept high, and folks can read/eat/sleep while ij traffic.

No, that's a constraint on infrastructure.

A taxi, robot or not, will never be able to meet peak demand... they'll be provisioned to meet 60-80% of peak demand and surge peak pricing. That's the key part of the robot car business model -- these companies want to soak you for using public infrastructure.

The Musk tunnel thing is totally different, it's an extreme version of Washington Beltway congestion pricing that charges you $3-60 to waste less time in express lanes.

Re: Waymo CEO dismisses Tesla self-driving plan: “This is not how it works”

#519

Post by invisible »

stefan_ wrote: post #476
invisible wrote: post #158

Earlier quoted context omitted.

> In the 4th quarter, we registered one accident for every 3.45 million miles driven in which drivers had Autopilot engaged. For those driving without Autopilot but with our active safety features, we registered one accident for every 2.05 million miles driven. For those driving without Autopilot and without our active safety features, we registered one accident for every 1.27 million miles driven. By comparison, NHT…

That doesn't tell the story. Autopilot is essentially forced off for anything but highway, so they are missing the most dangerous but are racking up millions of the least dangerous miles quick.

I can't actually find anything suggesting that the most dangerous roads are not highways. I really tried. Perhaps there are more small accidents on local roads, but they would tend to be less dangerous. The evidence suggests that highways are significantly more fatal[0].

Also, Autopilot is not essential forced off for anything but highways. That is just factually incorrect. Notably, straight-aways are where Autopilot works best and are the most dangerous by crashes.[1]

I found data suggesting that a lot of accidents happen because people rear-end stopped cars at intersections[2] and issues with navigating intersections[3]. That is an area where Autopilot is pretty great today.

0: https://www-fars.nhtsa.dot.gov/People/PeopleAllVictims.aspx (see last table)

1: https://www-fars.nhtsa.dot.gov/Vehicles/VehiclesAllVehicles.... (see last table)

2: https://www.hg.org/legal-articles/when-and-where-do-car-acci...

3: https://www-fars.nhtsa.dot.gov/Vehicles/VehiclesLocation.asp...

Re: Waymo CEO dismisses Tesla self-driving plan: “This is not how it works”

#520

Post by pas »

Spooky23 wrote: post #518
pas wrote: post #382

Earlier quoted context omitted.

Current peak demand in cars is limited by roads, and ... always will be. (Hence Musk trying to dig tunnels for more roads.) Which might be okay if throughput is kept high, and folks can read/eat/sleep while ij traffic.

No, that's a constraint on infrastructure. A taxi, robot or not, will never be able to meet peak demand... they'll be provisioned to meet 60-80% of peak demand and surge peak pricing. That's the key part of the robot car business model -- these companies want to soak you for using public infrastructure. The Musk tunnel thing is totally different, it's an extreme version of Washington Beltway congestion pricing that c…

If sitting in rush hour becomes even easier (and even more profitable, if you can work while going to a meeting), we'll have more of rush hour sitting.

It's enough if just ~20% of those who participate in traffic are doing so with a "self driving car", if they otherwise would have opted not to do so, we'll have even higher peaks.

Surge pricing might not matter much, because it's in the interest of the fleet operator/owner to get maximal utilization, to get maximum profit. After all Uber does surge pricing to motivate more drivers to come online, but if you have a constant supply (constant number of self-driving cars in the fleet), then you can simply use it to maximize profit. (Or to manage perception, ie. to show that yes there are cars available, sure, it comes at a high price, but they are "available".)

---

> these companies want to soak you for using public infrastructure

Yep, US cities are so burdened with problems (a nasty set of incentives) that public transit is non-viable, which drives even more subsidization of the sprawl, which then breeds these companies that try to "solve it". (I mean Uber/Lyft successfully increased the supply of taxis, which drove prices down. Yaay, great, but this just meant more congestion, more pollution, more sprawl, nothing fundamentally solved.) And of course at this point properly pricing externalities is an insurmountable political challenge, because people are so wedded to their way of life.

Post reply on HN