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

The Waymo World Model

waymo.com

Post reply on HN
Print view View original

691–699 of 699 posts

Re: The Waymo World Model

#691

Post by rogerrogerr »

fcantournet wrote: post #598
rogerrogerr wrote: post #517

Earlier quoted context omitted.

The goal is not to drive in all conditions; it is to drive in all drivable conditions. Human eyeballs also cannot see through dense fog clouds. Operating in these environments is extra credit with marginal utility in real life.

But humans react to this extremely differently than a self driving car. Humans take responsability, and the self-driving disengages and say : WELP. Oh sorry were you "enjoying your travel time to do something useful" as we very explicitely marketed ? Well now your wife is dead and it's your fault (legally). Kisses, Elon.

There’s nothing about the human reaction to a cloud of fog that can’t be replicated.

Re: The Waymo World Model

#693

Post by ActorNightly »

robotresearcher wrote: post #217
ActorNightly wrote: post #37

Earlier quoted context omitted.

The purpose of lidar is to prove error correction when you need it most in terms of camera accuracy loss. Humans do this, just in the sense of depth perception with both eyes.

Human depth perception uses stereo out to only about 2 or 3 meters, after which the distance between your eyes is not a useful baseline. Beyond 3m we use context clues and depth from motion when available.

Thats going off just the focal point.

We do a lot more internal image processing. For example, relative motion as seen by either eye helps improve accuracy by a whole lot, in the "medium" distance range.

Re: The Waymo World Model

#694

Post by nedt »

xnx wrote: post #671
doctoboggan wrote: post #326

Earlier quoted context omitted.

> they (1) don't actually charge enough maintain the variable cost of operations Why do you expect them to make money? Roads don't make money and no one thinks to complain about that. One of the purposes of government is to make investment in things that have more nebulous returns. Moving more people to public transit makes better cities, healthier and happier citizens, stronger communities, and lets us save money on…

This gets repeated a lot, but is unpersuasive. How much money should a transit system lose? $20 per trip? $40 per trip? There might be mass transit systems that make sense (e.g buses), but most mass transit in the US is terrible quality and a terrible value. One argument is that it's a jobs program for the disadvantaged, but even there we could find a lot of things more useful than moving around empty seats most of t…

It's not like only that the transit system is losing money? Every trip that's done with a car is also not fully paying for itself. We just keep ignoring how much hidden cost individual car rides have especially considering their use. Obviously heavier road users are even generating more costs, but they might have more use (like in delivering goods to a supermarket).

Re: The Waymo World Model

#695

Post by iamtheworstdev »

iknowstuff wrote: post #176
iamtheworstdev wrote: post #161

Earlier quoted context omitted.

he didn't refuse it. MobileEye or whoever cut Tesla off because they were using the lidar sensors in a way he didn't approve. From there he got mad and said "no more lidar!"

False. Mobileye never used lidar. Lmao where do you all come up with this

It's been a decade and it's hard to keep up with all of the drama and ego. It was the EyeQ3 vision system. It used cameras, radar, and ultrasonic sensors and Tesla was accessing them directly. MobileEye cut them off and Elon put his foot down and said "fine we'll just use crappy webcams and be fine."

Re: The Waymo World Model

#696

Post by senordevnyc »

ActorNightly wrote: post #658
senordevnyc wrote: post #232

Earlier quoted context omitted.

This is cool, but they are still not going about it the right way. This is legit hilarious to read from some random HN account.

I posted this before, but Ill post again - this is one of the few things I feel confident enough to say that most people in the space are doing wrong. You can save my post and reference it when we actually get full self driving (i.e you can take a nap in the backseat while your car drives you), because its going to be implemented pretty much like this: Humans don't drive well because we map vision policy to actions.…

when we actually get full self driving (i.e you can take a nap in the backseat while your car drives you)

What on earth? We already have this, it’s called Waymo. And the idea that they’re trying to catch up to Tesla is laughable.

Re: The Waymo World Model

#697

Post by schiffern »

essdas wrote: post #576
schiffern wrote: post #484

Earlier quoted context omitted.

>Waymo is doing the thing. This worldview is overly simplistic. Waymo has (very shrewdly, for prospective investors at least) executed a strategy that most quickly scales to 0.1% of the population. Unfortunately it doesn't scale further. The cars are too costly and the mapping is too costly. There is no workable plan for significant scale from Waymo. Tesla is executing the strategy that most quickly scales to 100% of…

> most quickly scales to 0.1% of the population. Unfortunately it doesn't scale further Data suggests that they’re already available to ~2% of the US population.

There's definitely not enough Waymos to replace the transport needs of 2% of the population, so 0.1% is a more accurate figure of merit.

Re: The Waymo World Model

#698

Post by rajnathani »

mooktakim wrote: post #110
mattlondon wrote: post #89

Suddenly all this focus on world models by Deep mind starts to make sense. I've never really thought of Waymo as a robot in the same way as e.g. a Boston Dynamics humanoid, but of course it is a robot of sorts. Google/Alphabet are so vertically integrated for AI when you think about it. Compare what they're doing - their own power generation , their own silicon, their own data centers, search Gmail YouTube Gemini wor…

Tesla built something like this for FSD training, they presented many years ago. I never understood why they did productize it. It would have made a brilliant Maps alternative, which country automatically update from Tesla cars on the road. Could live update with speed cameras and road conditions. Like many things they've fallen behind

Not really I think, they built a simulation engine for autonomous driving, for which tons of such exist out there including ones from Nvidia and also at least 1 open-source one. Using world models is different.

Re: The Waymo World Model

#699

Post by 0x457 »

runarberg wrote: post #688
0x457 wrote: post #687

Earlier quoted context omitted.

I live in LA and ride Waymo, subway and trains often (decided not buy a car after u-haul rear-ended me on a red light...). There is no way metro and trains will be a solid way of transportation in LA for at least a few more decades, even if people of Bel-Air stop fighting trains that go deep under Bel-Air and people of valley stop being unreasonable. Even if next president, rollbacks contingency requirement changes t…

> There are plenty of "dead end technologies" that we use to this day. Good point. However that does not apply to self driving cars because the problem it is trying to fix is a lot more complex then how to heat your microwave burrito more quickly. Self driving cars shares many of the same problems with human driven cars, including traffic congestion. If we see a mass employment of self driving cars in our cities, the…

> I think you are underselling your city here. LA has an excellent bus system which can be improved further still. LA also has the potential of installing a world class bicycle infrastructure, and me personally I am fairly optimistic that sometime in the near future, your city will put my city of Seattle to shame with your bike network.

No really, insanity that usually happens inside a bus in LA aside, it still goes on the same congested roads. "Bus Only" becomes a suggestion once rush hour starts (I can see bus only line backed up by not buses every day our of my window).

I mostly use metro because buses not worth it for me. Metro network is either wonderful stops or very bad stops. There is a massive "cleanup" attempt before the Olympics. Some will be right where you want to go and some are in the middle of a freeway 20 minutes of walking away from anything.

The fact that K line does not go through Getty Center to please a handful of people and car brains saddens me.

> LA also has the potential of installing a world class bicycle infrastructure

Please no. Cyclists are probably the most hated category on the road even by other cyclists. I'd ban cycling on public roads in most of LA if it were up to me.

Post reply on HN