News for hackers. Stuff that matters.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by Mo3 on Sunday March 16, 2025 @05:13PM (#43382373)
Aside of all the other obvious reasons to not get a Tesla these days this is #1 imo. Camera feeds and a neural network are not enough for self driving, no matter how much they're training. Never ever.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by nickthegreek on Sunday March 16, 2025 @05:19PM (#43382409)
At the very least they seem to have downsides that the can be easily overcame with a lidar system/combination of them. That alone is enough to help me decide when its my family that would be the passengers.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by crooked-v on Sunday March 16, 2025 @05:31PM (#43382534)
And every other modern auto-braking safety system, except for Subaru for some reason, incorporates at least basic proximity radar.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by bpodgursky on Sunday March 16, 2025 @05:21PM (#43382433)
I am not claiming that Tesla FSD is at this point, but it is obviously possible to use cameras and neural networks to drive a car, because that is literally how humans drive.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by seszett on Sunday March 16, 2025 @05:33PM (#43382551)
Do Tesla cars have stereo vision, though?

I don't think they build a 3D model of the world around them at all, while humans do (not only based on our stereo vision) and largely rely on that to drive.

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by sorenjan on Sunday March 16, 2025 @07:07PM (#43383441)
They do build a 3D model: https://youtu.be/6x-Xb_uT7ts?t=129

Although I think it's interesting that even in his demo there are cars popping in and out of existence in the 3D visualization of the environment. I don't think that makes much sense, if a car is observed and then occluded the logical conclusion should be that it's probably still there, maybe add an increasing uncertainty to its location, instead of assuming it disappeared.

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by bryanlarsen on Sunday March 16, 2025 @07:13PM (#43383490)
Tesla switched from a 3D model that with the popping glitchiness you describe to a 4D model without the popping several months ago.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by SR2Z on Monday March 17, 2025 @12:56AM (#43385336)
Sure, but during the entire time of Autopilot/FSD so far, Teslas have been crashing in ridiculous ways and killing their drivers.

The "but they've patched that failure!" excuse has ALSO been around, and while the cars are definitely getting better, they're still also definitely killing people in ridiculous ways.

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by ModernMech on Monday March 17, 2025 @02:53PM (#43391663)
Probably because they're not actually patching anything, but playing a shell game.

If they were serious about patching issues, they would have added LiDAR already. They're trying to software patch their way out of the fact their perception stack is essentially blind to huge classes of objects drivers encounter every day.

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by andsoitis on Sunday March 16, 2025 @05:41PM (#43382645)
Apparently, Tesla has stated that they do not use paired cameras for ranging or depth perception, which is the standard approach for stereoscopic vision.

Instead of stereoscopic vision, Tesla's neural net processes the data from the cameras to create a 3D view of the surroundings.

https://electrek.co/2021/07/07/hacker-tesla-full-self-drivin...

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by iknowstuff on Sunday March 16, 2025 @05:44PM (#43382692)
They do have 2-3 front facing cameras and their e2ee does necessarily get the same understanding of the world around it.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by rtkwe on Sunday March 16, 2025 @06:28PM (#43383130)
Neural nets generally don't have the same level of interconnections and neurons as the human brain because it's tougher to train. I agree in principal that the human brain should be possible to replicate in a computer but I'm not sure we can do it with the current design of single direction connections between nodes. I highly doubt we'll be able to scale it down to a machine that's reasonable to fit in a car any time soon though and that's what Tesla is promising they can do with the current hardware package installed in cars (never mind that they also promised it'd be possible with the last major revision and had to walk that back and will have to install upgraded electronics in people's vehicles, unless they're just going to strand all HW3 owners who paid of FSD but the hardware is to wimpy to handle it unless that's changed since the robotaxi event).
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by manquer on Sunday March 16, 2025 @05:40PM (#43382633)
The actual problem statement is can such a system do so safely. If it is just to drive a car all you need are some stepper motors and a raspberry pi.

The goal is to reduce accident and fatalities not eliminate jobs .

If LiDAR has 1/10th the fatality rate of camera setup and is less costlier than 10x the value of human life(as used in legal proceedings) then it still the only viable option

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by nickthegreek on Sunday March 16, 2025 @06:58PM (#43383385)
Don’t we want it to be like way better than humans at driving? The dream pitch was that we wouldn’t have tens of thousands of preventable deaths every year. So install the damn lidar. At some point it is coming down to penny pinching and that means that preventable deaths number will not sink to the promise.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by vel0city on Sunday March 16, 2025 @05:24PM (#43382469)
Humans have more senses than just vision. Also a few cameras aren't completely covering the range of human vision.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by intrasight on Sunday March 16, 2025 @05:30PM (#43382525)
Yes, but I don't think you could argue the point that the visual sense is probably 95% of it. But even so, it could be decades before computers achieve the sensory capabilities of the human visual system. Why not, during those hundred years, add some lidar to the sensory mix. Just because it didn't evolve in humans doesn't mean it's not a good thing. Bats and dolphins use the biological analogue very effectively.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by bcrl on Sunday March 16, 2025 @06:31PM (#43383151)
Vision isn't the only problem. Can computers do a decent job of theory-of-mind of other drivers? What about seeing a tire bouncing down off an overpass that's visible to the driver? Software developers love saying that it's the 95% that matters, but in safety critical systems, that simplification simply does not work.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by intrasight on Sunday March 16, 2025 @09:16PM (#43384244)
> Software developers love saying that it's the 95% that matters

Perhaps if they're early in their career. But any who have done challenging projects know that it's the last 5% that's the hardest and that often matters the most.

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by iknowstuff on Sunday March 16, 2025 @05:46PM (#43382710)
Their cameras have better dynamic range than humans, not to mention the neural net can consume multiple exposures simultaneously. From surround video.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by vel0city on Sunday March 16, 2025 @09:51PM (#43384445)
> Their cameras have better dynamic range than humans

This is obviously untrue if you've ever watched videos from sentry mode.

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by iknowstuff on Sunday March 16, 2025 @11:42PM (#43384989)
Sentry mode videos are sdr
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by aplummer on Sunday March 16, 2025 @07:47PM (#43383728)
Also humans suck at driving compared to what we expect of machines
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by bpodgursky on Sunday March 16, 2025 @05:27PM (#43382500)
? Do you use your sense of smell to drive a car? Are deaf people allowed to drive cars?

No, the cameras we have now, or at least the data processing, is probably not there yet, but it's absurd to claim it's "never" possible. It's obviously either possible in a year or fifteen years, all basic hardware is advancing fast.

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by tredre3 on Sunday March 16, 2025 @05:34PM (#43382559)
> it's absurd to claim it's "never" possible

Absolutely correct. The problem is that Elon has been saying that for ten years and we're not really closer to it being true.

So, in the meantime, would it kill him to admit that he was wrong about lidar?

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by vel0city on Sunday March 16, 2025 @10:35PM (#43384657)
> but it's absurd to claim it's "never" possible.

I've never made such a claim. I use self driving features in my car a good bit. They use more than just cameras though.

> Are deaf people allowed to drive cars?

Sure, but I'd say someone unable to hear things would be disadvanted at driving and potentially less safe. But you've also completely ignored the sense of feeling, understanding the feedback you're feeling on the wheel. Ignoring the feeling of the wheels on the road. Ignoring the feeling of the g forces as the car moves. Ignoring the feeling of body sway when it's windy.

You're also ignoring as others have mentioned the theory of mind of how other things actually behave around the streets. We're still not 100% sure these automated systems can actually comprehend these things.

All things a computer could sense, no doubt. But it's far more than just a few cameras.

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by bpodgursky on Monday March 17, 2025 @12:55AM (#43385333)
> Camera feeds and a neural network are not enough for self driving, no matter how much they're training. Never ever.

This is the OP claim I am saying is absurd.

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by jayd16 on Sunday March 16, 2025 @05:48PM (#43382735)
A human is not just a NN, but I do wonder how well a human could drive given the Tesla camera feed. Seems like it would surely be worse than behind the wheel.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by Aloisius on Sunday March 16, 2025 @05:41PM (#43382644)
I think artificial neural networks was implied and there is a world of difference between how biological and artificial neural networks function.

That said, I'm not sure that's what the barrier is. I think humans would have trouble driving with a fixed low resolution camera too.

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by Detrytus on Sunday March 16, 2025 @05:44PM (#43382689)
1. Human "neural network" is few orders of magnitude more powerful than the best neural network available right now.

2. This might be obvious question but: Humans have two eyes and they are few centimeters apart so that we could use stereoscopic vision to estimate the distance. Does Tesla FSD even attempt to do something similar?

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by bpodgursky on Sunday March 16, 2025 @05:47PM (#43382729)
It is legal and safe to drive with 1 working eye.

Humans have many ways of modeling distances in the real world that do not rely on stereoscopic depth.

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by beepbooptheory on Sunday March 16, 2025 @05:42PM (#43382666)
Hubert Dreyfus rolling in his grave..
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by ryandvm on Monday March 17, 2025 @09:48AM (#43388586)
The problem with that argument is that a Tesla isn't nearly as clever as a human. I have never thought a white tractor trailer disappeared as it crossed in front of me against a bright sky. I know that I should drive a little bit slower on Halloween evening because there are kids running around. I know that I need to be a little more cautious driving past the playground or dog park. I have object permanence.

As it is, AI just isn't smart enough to rely on only vision and they ought to be taking advantage of every economically viable sensor to make up for the difference. Hell, I thought the goal was to do better than humans?

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by stnmtn on Monday March 17, 2025 @12:49PM (#43390420)
I get your point, but do you think, over the course of a decade, the average human driver or the average car with tesla FSD is more likely to have an accident where the the fault is their own?
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by ModernMech on Monday March 17, 2025 @10:05AM (#43388745)
Eyes are not cameras and brains are not neural networks. Until we have artificial techniques that can match the performance and functionality of this amazing biological equipment we have, our cars must rely on other sensors like LiDAR to make up for the deficiencies in our artificial perception networks that are not overcome by current techniques.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by consumer451 on Monday March 17, 2025 @05:05PM (#43392737)
Musk himself said during the last Tesla AI Day: ~"FSD will basically require AGI."

I would add that it would require not only AGI, but real-time AGI!

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by sMarsIntruder on Sunday March 16, 2025 @05:38PM (#43382615)
You started with a bias and ended with another one.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by dist-epoch on Sunday March 16, 2025 @05:49PM (#43382757)
> Camera feeds and a neural network are not enough for self driving

I guess we should ban humans driving then.

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by sMarsIntruder on Monday March 17, 2025 @02:04AM (#43385582)
I’m noticing how comments with common sense gets easily downvoted.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by desixavryn on Sunday March 16, 2025 @05:45PM (#43382707)
Have you used latest FSD on HW4 recently? If not, please try it out for a few days and then come back to correct your comment :-)
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by Hamuko on Sunday March 16, 2025 @05:50PM (#43382762)
Have they managed to fix the "parking sensors"?
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by desixavryn on Sunday March 16, 2025 @06:07PM (#43382934)
Using vision only on my HW4-Model-Y it seems to work fine.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by windward on Monday March 17, 2025 @06:21AM (#43386841)
Honest question: why does the HW4 matter? Hasn't FSD been sold as a service all cars with older hardware can use?
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by lowmagnet on Sunday March 16, 2025 @08:32PM (#43383986)
I do wonder if this is provable via information theory.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by hnburnsy on Monday March 17, 2025 @12:11AM (#43385141)
Agreed, but I believe that camera feeds and infrastructure changes could be enough.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by FeloniousHam on Monday March 17, 2025 @09:31AM (#43388417)
Have you actually used FSD for, say, a week?

I'm a fan and everyday user of FSD. It's not perfect, but it's immensely useful, and frankly amazing. It works. It drives me to work and back everyday, with maybe 1 or 2 interventions each way.

I've never met any actual Tesla driver who was confused by the marketing (they're all tech guys). Some like it, some don't think it's worth the money, but everyone understands what it is.

(I'm not arguing against more detection hardware, but engineering is about trade offs.)

Re:Tesla drives into Wile E. Coyote fake road wall in camera... by bingofuel on Monday March 17, 2025 @12:22PM (#43390129)
But not every Tesla owner is a tech guy. Not every Tesla driver lives in perfect sunny conditions. And in this case, a misinformed (or not informed at all) buyer's "engineering trade off" is death.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by thefourthchime on Wednesday March 19, 2025 @12:02PM (#43413767)
It drives at night and in pretty heavy rain just fine.
Re:Tesla drives into Wile E. Coyote fake road wall in camera... by FeloniousHam on Monday March 24, 2025 @02:56PM (#43464270)
It is not fully autonomous. It quits when blinded by the rising sun.

It does drive in the rain. I just drove for an hour through a torrential downpour on a road trip, way less stressful than without it.