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

Cruise is opening driverless cars to the public in San Francisco

getcruise.com

Post reply on HN
Print view View original

551–560 of 652 posts

Re: Cruise is opening driverless cars to the public in San Francisco

#551

Post by konschubert »

kvogt wrote: post #93
walrus01 wrote: post #8

Based on this image: https://images.ctfassets.net/95kuvdv8zn1v/6h1C7lPC79OLOlddEE... They and their VC backers are clearly betting on the concept that radars + lidar + imaging will be the ultimate successful solution in full self driving cars, as a completely opposite design and engineering philosophy from Tesla attempting to do "full self driving" with camera sensors and categorical rejection of lidar. It is interes…

Cruise CEO here. Our strategy has been to solve the challenges needed to operate driverless robotaxis on a well-equipped vehicle, then aggressively drive the cost down. Many OEMs are doing this in reverse order. They're trying to squeeze orders of magnitude of performance gains out of really low-cost hardware. Today it's unclear what strategy will win. In a few years our next generation of low-cost compute and sensin…

As a bike rider, father, and sometimes inattentive human, I would like to say thank you for the safety you are bringing to our cities.

Re: Cruise is opening driverless cars to the public in San Francisco

#552

Post by jhbadger »

toss1 wrote: post #345
astrange wrote: post #329

Earlier quoted context omitted.

Theranos didn't use the normal set of VCs because they all thought it was a scam; they raised from some random rich people who weren't professional tech VCs instead. It's unfortunate the only thing she was convicted for was defrauding them, since being accredited investors they should be able to live with that. As for Elon, he's currently doing a bad boy anti-government bit in an attempt to make Tesla "electric cars…

>> It's unfortunate the only thing she was convicted for was defrauding them, since being accredited investors they should be able to live with that. NO. Investors are supposed to be able to live with all the usual risks of technology, execution, marketplace dynamics, etc. They are NOT supposed to be OK with deliberate fraud. If you invest at Early_Round when the tech looks promising, but then it fails to develop, CE…

But I think the point is that she wasn't convicted of endangering people's lives, which many would consider a far greater crime than just a con defrauding some gullible marks. People depended on those tests. They made choices (such as whether or not to have surgeries) based on the results.

Re: Cruise is opening driverless cars to the public in San Francisco

#553

Post by audunw »

mhh__ wrote: post #534
ivalm wrote: post #482

Earlier quoted context omitted.

We know in a sense that very simple animals do it and it doesn’t require decision making (in a sense that LiDAR only helps with perception).

So I can teach my dog to drive the car?

Your dog can reliably detect objects, judge distance and avoid them.

That's all the person is saying. Simple animals can use only vision to do what we're using lidar and radar to do. But neither camera, lidar nor radar or any combination of them guarantees that you'll be able to make a computer drive a car in all situations.

For me, intuitively, the problem of reconstructing a distance field from cameras must be way harder than say, trying to predict what a person on a bike will do next, or detect road lines on the road in heavy rain or snow. So it seems very likely that an "AI" capable of driving a car in all situations would be powerful to not need lidar or radar (though I don't see the point of dropping radar, as it gives you some ability to "see" around objects which can make cars better than humans)

Re: Cruise is opening driverless cars to the public in San Francisco

#554

Post by CPLX »

astrange wrote: post #329
drzaiusapelord wrote: post #328

Earlier quoted context omitted.

Game theory and personality dynamics strategy applies to C-levels too! For some industries, companies, etc you want to be a Holmes type of total confidence. It depends who your market is, and by that I mean the VC's you want to attract. If they go for the "arrogant boy/girl genius" schtick then that's what you do. If they want a humble intellectual, then that's what you do instead. Conversely, you may alternate betwe…

Theranos didn't use the normal set of VCs because they all thought it was a scam; they raised from some random rich people who weren't professional tech VCs instead. It's unfortunate the only thing she was convicted for was defrauding them, since being accredited investors they should be able to live with that. As for Elon, he's currently doing a bad boy anti-government bit in an attempt to make Tesla "electric cars…

Tim Draper wasn’t random.

Re: Cruise is opening driverless cars to the public in San Francisco

#555

Post by sam_goody »

silverlake wrote: post #260
kvogt wrote: post #93

Earlier quoted context omitted.

Cruise CEO here. Our strategy has been to solve the challenges needed to operate driverless robotaxis on a well-equipped vehicle, then aggressively drive the cost down. Many OEMs are doing this in reverse order. They're trying to squeeze orders of magnitude of performance gains out of really low-cost hardware. Today it's unclear what strategy will win. In a few years our next generation of low-cost compute and sensin…

Tesla limited themselves to cameras because Musk said “humans can do it with two eyes”. He also didn’t like the look of LiDAR on cars. Such an idiotic decision. Good to see Cruise is not lead by a mega-maniacal CEO.

A more accurate summary of Tesla's position is that they beleive that the incoming data from different systems (lidar, radar, visual, etc) must be merged and very often there is contradictory data.

Resolving that correctly takes time (in ms), adds complexity and will sometimes be incorrectly judged.

Since the visual data is the more accurate the vast majority of the time, it will anyways take precedence over the other input. As humans have proven that visual is technically enough, they decided it makes more sense to squeeze the most out of the visual, rather than collecting other data, crunching it, then (in most cases) discarding it.

I am not sure they are right, and am pretty sure that even if so - they need better cameras.

But misquoting them doesn't really help your argument.

Re: Cruise is opening driverless cars to the public in San Francisco

#556

Post by hngpt2 »

walrus01 wrote: post #8

Based on this image: https://images.ctfassets.net/95kuvdv8zn1v/6h1C7lPC79OLOlddEE... They and their VC backers are clearly betting on the concept that radars + lidar + imaging will be the ultimate successful solution in full self driving cars, as a completely opposite design and engineering philosophy from Tesla attempting to do "full self driving" with camera sensors and categorical rejection of lidar. It is interes…

[dead]

Re: Cruise is opening driverless cars to the public in San Francisco

#557

Post by andy_ppp »

walrus01 wrote: post #391
gooseus wrote: post #264

Earlier quoted context omitted.

"This is a solved problem.", "We're light-years ahead of the competition", or "We expect everyone with existing hardware to be able to monetize their cars as robo-taxis by the end of next year" are other great examples.

to be fair, this is because Elizabeth Holmes is probably a sociopath, so you'd expect to hear something overly confident like that from such. Not that we knew it at the time.

I've met quite a few people like her and I really don't believe they are sociopaths - they say things that sound real and even convince themselves of it but as soon as you start peeling back details they don't exist. It's like you are talking to a Turing machine and it's you failing the test. People on these forums never really speak to these people because we are already largely scientists and love empirical evidence and work with and hang around similar people. There's a whole class of people who have learned to pattern match talking like us without anything to back it up. See the industry "Comms" for many many examples.

Re: Cruise is opening driverless cars to the public in San Francisco

#558

Post by olau »

YZF wrote: post #500
LegitShady wrote: post #371

Earlier quoted context omitted.

It was a discussion about how such a system must function with all the different inputs vs the vision only model. If you have radar and lidar and vision, then you have at least three different specialist machine learning models running, and then another model running that takes their outputs and decides what the car is going to do. You may have even more than that, some doing specific tasks like localization. Neural…

That argument doesn't make any sense though. Clearly radar + lidar + vision is a superset of vision only. It can perform as well as vision only if you disable the other two. So the claim is there is absolutely no scenario where the other systems can contribute which seems to be false. E.g. Tesla's cameras are blinded pointing straight on into the sun, the car even tells you the cameras are blinded. If the cameras are…

> So the claim is there is absolutely no scenario where the other systems can contribute which seems to be false.

No. The claim was that pouring resources into those other systems is better spent improving the vision system, within the context that Tesla and Comma.ai is operating.

> But clearly having more data can't make things worse.

There are several examples of cars hitting stuff because one part of the sensor suite thought there was a problem, or that there wasn't a problem.

In general, the more complex you make a system, the more complex the failure modes get.

Re: Cruise is opening driverless cars to the public in San Francisco

#559

Post by zppln »

kvogt wrote: post #93
walrus01 wrote: post #8

Based on this image: https://images.ctfassets.net/95kuvdv8zn1v/6h1C7lPC79OLOlddEE... They and their VC backers are clearly betting on the concept that radars + lidar + imaging will be the ultimate successful solution in full self driving cars, as a completely opposite design and engineering philosophy from Tesla attempting to do "full self driving" with camera sensors and categorical rejection of lidar. It is interes…

Cruise CEO here. Our strategy has been to solve the challenges needed to operate driverless robotaxis on a well-equipped vehicle, then aggressively drive the cost down. Many OEMs are doing this in reverse order. They're trying to squeeze orders of magnitude of performance gains out of really low-cost hardware. Today it's unclear what strategy will win. In a few years our next generation of low-cost compute and sensin…

Will you share your systems safety work? How many fatalities per drive hour do you expect?

Re: Cruise is opening driverless cars to the public in San Francisco

#560

Post by silentsea90 »

pm90 wrote: post #526
silentsea90 wrote: post #506

Earlier quoted context omitted.

I don't know the answers. They're definitely worth figuring out, but the legal questions do not take away from the sheer gall to dream and achieve. Figuring out liability is in the domain of law, which is easy compared to building such engineering marvels. I am sure we can figure it out. Self driving car companies take safety VERY VERY seriously not just for potential legal consequences but because a single incident…

None of the other situations you’ve mentioned are similar to a self driving car being unleashed on an unsuspecting dense urban area without setting clear rules for what’s to happen in the worst case. These are the frameworks which guide decision making. Without clear guidelines, I have a lot of reason to doubt that eg a product team that really wants to deliver a feature will ship something without the proper amount…

I don't claim that laws don't have catching up to do, or that Cruise shouldn't be liable for any loss to life or property they may cause. We can be appreciate their achievement while being cognizant of the need for laws around this.

My second comment was targeted at naysayers and pessimists who find faults and issues everywhere. Yes gaps exist and that's how tech and society plays catch up, but let's not miss the forest for the trees here.

Post reply on HN