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

Inside Waymo's Secret World for Training Self-Driving Cars

theatlantic.com

Post reply on HN
Print view View original

81–90 of 146 posts

Re: Inside Waymo's Secret World for Training Self-Driving Cars

#81

Post by skgoa »

nopinsight wrote: post #19

The fact that Waymo revealed their "secret" tools for advancing this crucial technology implies that either: 1) They believe no one can quite catch up before they can launch the technology. Since they know several competitors have huge resources and brilliant people, it means they are quite close to launch. 2) These tools have become open secret within the industry, so no harm is done to their competitive position by…

There was nothing secret in that article. Everyone is doing it. E.g. BMW won an industry innovation award for their simulation work years ago. Yes, right around the time when Google were still just driving around in a parking lot.

Re: Inside Waymo's Secret World for Training Self-Driving Cars

#82

Post by wastedhours »

peoplewindow wrote: post #63
dragonwriter wrote: post #61

Earlier quoted context omitted.

> How much money does Google Photos make? Last time I used it, there were no ads in it. Google Photos is, in a sense, a big ad with prominent calls to action for selling storage space. > Isn't speech recognition nearly at human levels of comprehension, at least for US English and outside of highly technical jargon? Recognition, maybe (though I think no). Applying recognized text to get the desired outcome, not even c…

But selling disk storage is a very marginal business and few people take huge numbers of photos they value enough to pay for storage of (vs sticking them on Facebook or Instagram which compresses the hell out of them but does it for free). I really doubt this is a significant business for Google.

But this will only compound, no? If people keep using it, the switching "cost" (cognitive/time) increases exponentially and people will near the storage limits.

Re: Inside Waymo's Secret World for Training Self-Driving Cars

#83

Post by agumonkey »

Animats wrote: post #46

There's a lot of misunderstanding about self-driving. Mostly because nobody is publishing much. If you want to do it right, you start with geometry. The first step is capturing range imagery and grinding it down to a 3D model of the world. This tells you where you physically can go. That's where we were at the DARPA Grand Challenge over a decade ago. Then comes moving object popout. What out there isn't a stationary…

Is there a model ponderated by momentum ? Decisions have different consequences if you're at 30 vs 50mph.

Re: Inside Waymo's Secret World for Training Self-Driving Cars

#84

Post by notahacker »

iUsedToCode wrote: post #53
bit_logic wrote: post #48

Self driving cars is based on machine learning which is basically processing massive amounts of past data. This is great for routine situations and it seems Waymo is making good progress covering most of these. However the key weakness is the lack of true intelligence. When anything unusual or unexpected happens, the best it can do is simply safely shutdown and wait for a human to intervene. And the car can't really…

There arent that many novel scenarios on the road. Sure, Google can't program around the possibility of an airplane falling down on you, but how often does that happen? It doesn't have to be perfect. Just very good and improving. Some time ago google shared a gif of a wheelchair chasing a duck in the middle of the road. The car didn't understand it, so it just stopped. Good enough for me. Obviously, they have a lot o…

Novel scenarios might not be common compared with miles of traffic-following drudgery, but even really bad human drivers deal with novel scenarios on the road more often than they have accidents.

Stopping might be a sensible safety protocol in some situations, but it isn't in others (not to mention the situations where the car may stop too late because it doesn't actually recognise that a novel scenario is about to occur).

So if you want Level 5 driving, and not just a very impressive demo which is safe provided a human watches it attentively enough to take emergency measures and is able to override it when it decides it can't process a situation well enough to proceed, you need the AI to be pretty damn close to perfect in its judgement of how to react to a huge number of edge cases.

Re: Inside Waymo's Secret World for Training Self-Driving Cars

#85

Post by tim333 »

user5994461 wrote: post #71
pgodzin wrote: post #28

Earlier quoted context omitted.

Aren't those examples in which it'd be fine for it to be confused? That confusion shouldn't lead to any dangerous situations, just a different speed than other cars. It would also be interesting to know whether Waymo factors in other cars' behavior. If everyone is going 80 while the speed limit is 65, will it factor that in and go 80 or keep with the speed limit?

>>> If everyone is going 80 while the speed limit is 65, will it factor that in and go 80 or keep with the speed limit? No, that's illegal to drive over the speed limit.

Perhaps. Tesla allow the driver to get the system to exceed the limit. https://www.theguardian.com/technology/2017/jan/16/tesla-all...

Re: Inside Waymo's Secret World for Training Self-Driving Cars

#86

Post by ghshephard »

user5994461 wrote: post #71
pgodzin wrote: post #28

Earlier quoted context omitted.

Aren't those examples in which it'd be fine for it to be confused? That confusion shouldn't lead to any dangerous situations, just a different speed than other cars. It would also be interesting to know whether Waymo factors in other cars' behavior. If everyone is going 80 while the speed limit is 65, will it factor that in and go 80 or keep with the speed limit?

>>> If everyone is going 80 while the speed limit is 65, will it factor that in and go 80 or keep with the speed limit? No, that's illegal to drive over the speed limit.

Google's driverless cars designed to exceed speed limit

See: http://www.bbc.com/news/technology-28851996

Re: Inside Waymo's Secret World for Training Self-Driving Cars

#87

Post by ghaff »

jacquesm wrote: post #8
altotrees wrote: post #2

Clearly the progress being made in the area of self-driving cars is undeniable. Seeing the articles and discussions crop up on a daily basis are making me wonder if we are headed for an AI Winter scenario in this area within the next decade or if this is the real deal and we will see self-driving cars at dealerships within 20 years. This is so far from being my area of expertise. Just one observer's thoughts/question…

> Seeing the articles and discussions crop up on a daily basis are making me wonder if we are headed for an AI Winter scenario in this area within the next decade or if this is the real deal and we will see self-driving cars at dealerships within 20 years. It will be either one of those two, if it doesn't happen in the next 20 years I would bet against it happening at all until AGI is a reality, assuming it will ever…

For a lot of people, true hands-off autonomous driving on limited access divided highways is by itself a huge convenience feature and, potentially, a huge safety feature. And this seems like relatively low hanging fruit. People will pay for this. (And are, in a limited way, on more and more cars today.)

It's not sexy and doesn't fulfill autonomous taxis everywhere fantasies but for the vast majority of people who don't live in or near urban cores it would likely be a significant quality of life improvement.

Re: Inside Waymo's Secret World for Training Self-Driving Cars

#88

Post by tim333 »

peoplewindow wrote: post #56
Animats wrote: post #6

Now this is doing it right. That's how car companies do it. Big test tracks, and years of test track time, plus simulation and test rigs. As the article points out, Google has far more autonomous driving miles than everybody else put together. They're on the hard cases, too; not just driving on freeways.

Well, do they have hard cases? I was surprised to read in the article that they were flummoxed by roundabouts, of all things. And that they were apparently surprised to discover that multi-lane roundabouts even exist at all. Have none of these engineers ever gone on holiday to Europe? If they think a multi-lane roundabout is an advanced case, what will they make of this? https://www.wired.com/2016/08/brilliant-sorcer…

And at least the Swindon one has markings. Paris has the same kind of thing without https://www.youtube.com/watch?v=-2RCPpdmSVg

Re: Inside Waymo's Secret World for Training Self-Driving Cars

#89

Post by dsfyu404ed »

pgodzin wrote: post #28
stevenwoo wrote: post #25

Earlier quoted context omitted.

I occassionally come across construction that is marked in an unusual way - at 65 miles an hour when the speed limit is 75. Also, when they are changing freeways and restriping them (leaving the original old faded lanes and before new lines go up) sometimes it's confusing to me and the traffic is flowing at 80 miles an hour, IIRC this was happening on 880 for a while about 10 years back. If the AI is confused by a ro…

Aren't those examples in which it'd be fine for it to be confused? That confusion shouldn't lead to any dangerous situations, just a different speed than other cars. It would also be interesting to know whether Waymo factors in other cars' behavior. If everyone is going 80 while the speed limit is 65, will it factor that in and go 80 or keep with the speed limit?

A slower speed is dangerous to everyone. One car slows down and everyone else needs to get around them. That results in people who would otherwise be content to flow with traffic making aggressive moves to pass the obstruction. These actions catches some people in the other lanes unaware and some of them insensitively hit their brakes (kind of like the people who brake for curves at low speed because "it's what you do") which exacerbates the problem.

Now add in a poorly marked road dense traffic and construction. Other drivers (as a ground) don't really have the spare attention to allocate to the fallout of someone who's slowing down.

edit: This is backed up by traffic studies When everyone is driving the same speed things are safer because there's less "friction" (I think that's the correct term) between traffic elements.

Re: Inside Waymo's Secret World for Training Self-Driving Cars

#90

Post by dsfyu404ed »

jooke wrote: post #64
flamedoge wrote: post #57

> But Peng also presented the position of the traditional automakers. He said that they are trying to do something fundamentally different. Instead of aiming for the full autonomy moon shot, they are trying to add driver-assistance technologies, “make a little money,” and then step forward toward full autonomy. It’s not fair to compare Waymo, which has the resources and corporate freedom to put a $70,000 laser range…

Who's going to buy a car that might tell them they can't drive?

rich parents will buy them as high-school graduation presents. It's a very, very tiny market.
Post reply on HN