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

441–450 of 652 posts

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

#441

Post by ck2 »

I still want to know who goes to prison when a pedestrian or cyclist is killed.

Because no-one is arguing someone will be killed, only that "human drivers kill people too" or "machines will kill fewer, eventually".

Will it be the passenger? The coder for bad code? The person who assembled the hardware for improperly tested malfunctioning sensors? The person who made the malfunctioning sensors?

Who? Because someone is responsible. You can't just spread the blame around and the corporation just pays a fine and shrugs after someone's life is ended.

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

#442

Post by amacneil »

justapassenger wrote: post #321
amacneil wrote: post #300

Earlier quoted context omitted.

Tesla's approach is a textbook case of differing constraints. Despite Elon's commentary, the reason Tesla does not use lidar is sensor cost (waymo/cruise cars probably have 250k+ worth of sensors on them), and reason Tesla does not use radar is supply chain.

Sensor cost is a lame excuse, especially coming from a company that's one of the leaders of EV world. Battery prices are decreasing, but sensors wouldn't? Tesla put constrain on themselves, because they wanted to sell product years before they existed, and they were super cheap on HW, to optimize their costs. They optimized their solution, before they even had a slightest idea what's needed to make it work.

I don't think it's a lame excuse at all. Tesla barely managed to get a profitable car on the market given battery costs, and had to bet their company on a strategy of massive investment and reducing costs over a decade+ period.

They are also ahead of every other auto maker in commercially purchasable L2 autopilot.

With that context, it doesn't make sense to add $50k+ to the price of their cars (e.g. doubling the cost of a Model Y) just to get slightly better autopilot performance. Lidar would help them in some cases (e.g. stationary objects), but it's not a panacea. Their strategy makes perfect sense given they want to sell semi-affordable cars to the public.

On the flipside, Cruise & Waymo's strategy of "geofenced L4 at any initial cost" makes perfect sense for their short term robotaxi ambitions.

Maybe in ~10 years these strategies will intersect, but for the time being they are completely different products.

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

#443

Post by wutbrodo »

rfrey wrote: post #439
BurningFrog wrote: post #420

Earlier quoted context omitted.

I've given it some thought, and I think the SDC manufacturer must be liable for any accidents the SDC causes. Who else is there? The passengers certainly can't be responsible for any programming or manufacturing errors. There are corner cases and exceptions, but that has to be the rule. Which should mean that as and SDC owner, you don't have to pay car insurance.

>the SDC manufacturer must be liable for any accidents the SDC causes. There's the rub... How do you seperate those from the rest?

We already have a system that relies on assigning fault among the multiple parties involved in an accident. This same approach applies just as well to SDC accidents. It would be even easier than the status quo, given the much richer data SDCs could be regulatorily mandated to provide.

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

#444

Post by robotresearcher »

jowday wrote: post #436
upbeat_general wrote: post #413

Earlier quoted context omitted.

I think the comparison comes from the fact that humans infer a 3D map from stereo vision whereas LiDAR to some extent gives you the ground truth 3D map. You’re right that it falls apart pretty quickly though.

Except Tesla isn’t inferring a 3D map from stereo vision either, at least not outside of the front facing cameras - they’re using monocular depth prediction.

Neither are humans. Our eyes are so close together there's almost no disparity between the eye images beyond a handful of meters. We do 3D by inference beyond the near-field.

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

#445

Post by Retric »

jowday wrote: post #435
Retric wrote: post #398

Earlier quoted context omitted.

Novelty isn’t nearly as big of a problem as you might think. One of Wamo’s famous videos was someone on an electric scooter chasing a duck in the middle of the street. That’s very odd behavior, but the car followed the rather simple option of just not hitting them and going forward when possible. Cars really don’t need to identify what something is just it’s location and movement which is a vastly easier problem. A t…

> Cars really don’t need to identify what something is just it’s location and movement which is a vastly easier problem. A trash can rolling down the street can be treated just like an oil drum doing the same thing etc. You’d think that, until you encounter something like a turn restriction sign with a bizarre conditional restriction that it’s never seen before. At which point the car needs to OCR the text, parse the…

Or treat that turn restriction as applying 100% of the time.

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

#446

Post by sangnoir »

zaptrem wrote: post #338
xp84 wrote: post #302

Earlier quoted context omitted.

> "driver might manage all driving duties on surface streets then become a passenger as the car enters a highway." This is the kind of setup I can't wrap my head around. The car might "require" you to take over when you exit the highway, but it can't exactly "make" you. If you fall asleep on the freeway and the car isn't willing/able to drive at the end of your journey, or in edge cases if you were to pass out, etc.…

Pull into breakdown lane.

What happens when there isn't one? Roadworks, and accidents cause frequent closures of the breakdown lane. L3 has a lot of edge cases where the vehicle is supposedly too dumb to drive, but smart enough to know it shouldn't drive. It may be death by a thousand cuts.

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

#447

Post by robotresearcher »

jowday wrote: post #438
robotresearcher wrote: post #422

Earlier quoted context omitted.

The similarity is in the problem to be solved, not the details of the compute pipeline. Depth must be inferred somehow, rather than measured by actively interacting with the surface, as it is in LiDAR.

If you really wanted to make this argument, you wouldn’t even want to bother inferring depth, since that’s not what humans do, not directly at least. If you’re actually trying to obtain a depth map as part of your pipeline, LIDAR (or LiDAR + vision feeding into a denser depth prediction model) would always be a better strategy, cost aside, since determining depth from images is an ill posed problem.

My claim is that humans use their eyes as a primary input for driving. I don't think it's controversial. We don't let eyeless people drive. Eyes do not shoot out lasers.

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

#448

Post by wutbrodo »

gibsonf1 wrote: post #207
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…

Unless you have a conceptual AI with causal systems understanding reacting in real-time to a spacetime model of the world based on current and recent events, people are going to get injured and or killed by unusual real-world events riding in these autonomous cars. Although cats and dogs have great perception, we don't let them drive our cars for a reason.

> people are going to get injured and or killed by unusual real-world events riding in these autonomous cars

I don't think anyone inside or outside of the AV industry is expecting that there will be zero injuries or fatalities involving AVs. Why would that be the bar, when AV rides displace human drives that already injure and kill tens of thousands?

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

#449

Post by willcipriano »

ironman1478 wrote: post #306
julienb_sea wrote: post #253

Earlier quoted context omitted.

This problem should be solvable in software. The car can simply refuse to operate in a situation where maintenance is required.

How do you know maintenance is required in a completely automated fashion?

Put a rfid tag in the tire, store how many rotations that tire takes over time. Once it reaches a threshold, refuse to spin that tire further.

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

#450

Post by wutbrodo »

belter wrote: post #187
pulse7 wrote: post #155

Earlier quoted context omitted.

Thank you for sharing! Sharing details on how your system works brings confidence to customers...

Did you watch the videos? Based on the reply to the question: What sets Cruise technology apart from others like Waymo, Tesla...In other words, how was this difficult technical problem, solved in a way others were unable to do so far... And whose reply you can hear here (video at the correct time): https://youtu.be/ABto5nqWgc0?list=PLkK2JX1iHuzz7W8z3roCZEqML... Thank you, but wont be volunteering to ride one of these…

Lol, what answer do you expect from a question like that? At a high level, Cruise is taking the same approach as Waymo, and a different one from Tesla: start with lots of hardware, HD maps, and a targeted operating domain, then try to scale. Answering in any more detail would be a) giving away trade secrets and b) rely on knowing trade secrets about Waymo's cars that they probably don't know.
Post reply on HN