Live data from Hacker News

Tesla in self-drive mode slams into police car in Orange County

ktla.com

81–90 of 95 posts

Re: Tesla in self-drive mode slams into police car in Orange County

#81
post #16

Earlier quoted context omitted.

I hope there are also repercussion for whoever named this tech "full self driving"

There are, I hear he just got a $56bn dollar bonus. Worth every penny that Elon, his work is so fine and free from flaw. Sigh

A bonus that shareholders previously approved for very concrete goals that he has met.

Re: Tesla in self-drive mode slams into police car in Orange County

#82
post #79

Earlier quoted context omitted.

>I wonder if it’s prone to hallucinations like its LLM cousins. When was their vision system not glitching? A year ago it was seeing the moon on the sky as a perpetual yellow traffic light.

That's not hallucination, that's image recognition issue.

How do you define the border between hallucination vs recognition issue?

Re: Tesla in self-drive mode slams into police car in Orange County

#83
post #79

Earlier quoted context omitted.

That's not hallucination, that's image recognition issue.

How do you define the border between hallucination vs recognition issue?

Isolating an object is baseline recognition. Any sort of semantic association is baseline interpretation. So that border is somewhere between "light source" and "traffic light".

Re: Tesla in self-drive mode slams into police car in Orange County

#84

It boggles my mind that any self driving system would knowingly drive into any object, no matter what it is, as large as a police car. I'd be willing to bet that Elon's insistence on only using cameras is combining with the flashing lights to effectively make objects that appear in some frames of video, and not in others, as the camera recovers from the flash... and this makes it appear as noise, and below a certain…

The problem is that Tesla uses an ML classifier to identify objects it sees. Once it has identified them, it can estimate how big it is, and thus how far away it is. But if the classifier doesn't identify the object, it might as well not exist. Contrast this with Subaru, who uses binocular view through two widely spaced cameras to build a depth map of the view in front of it. It doesn't know or care what is there, on…

This is false. Tesla has an occupancy network.

Re: Tesla in self-drive mode slams into police car in Orange County

#85
post #84

Earlier quoted context omitted.

The problem is that Tesla uses an ML classifier to identify objects it sees. Once it has identified them, it can estimate how big it is, and thus how far away it is. But if the classifier doesn't identify the object, it might as well not exist. Contrast this with Subaru, who uses binocular view through two widely spaced cameras to build a depth map of the view in front of it. It doesn't know or care what is there, on…

This is false. Tesla has an occupancy network.

Please pardon my failure to remember the correct term. But fundamentally, it constructs a 3d model based on what it thinks each pixel represents. And sometimes gets it wrong, leading to it having a pattern of driving into stationary objects.

Re: Tesla in self-drive mode slams into police car in Orange County

#86
post #84

Earlier quoted context omitted.

This is false. Tesla has an occupancy network.

Please pardon my failure to remember the correct term. But fundamentally, it constructs a 3d model based on what it thinks each pixel represents. And sometimes gets it wrong, leading to it having a pattern of driving into stationary objects.

It's not a matter of terminology. Tesla's occupancy network "doesn't know or care what is there, only that something is there it shouldn't hit". As the name suggests, it's solely concerned with what space is occupied and what space is unoccupied, not with identifying objects. As such, the distinction you tried to draw between Subaru and Tesla is false, no matter what terms you use.

Re: Tesla in self-drive mode slams into police car in Orange County

#87
post #75

Earlier quoted context omitted.

> redesign the driving system from scratch to work on silicon Nobody proposed this. My Subaru can lane keep even in whiteout conditions (if there is a car in front of me) due to radar. I can’t do that. Tesla’s can’t do that. > it will occasionally do something as stupid looking as "drive into a police car with flashing lights on" Two sensors operating on different wavelengths can add more information than a single do…

Huh? The Subaru EyeSight system uses only cameras. There is no radar. https://www.subaru.com/eyesight.html IME it works pretty well but cuts out in heavy precipitation because the cameras don't get a clear image.

EyeSight is not the only lane-keeping component.

Re: Tesla in self-drive mode slams into police car in Orange County

#88
post #75

Earlier quoted context omitted.

Huh? The Subaru EyeSight system uses only cameras. There is no radar. https://www.subaru.com/eyesight.html IME it works pretty well but cuts out in heavy precipitation because the cameras don't get a clear image.

EyeSight is not the only lane-keeping component.

EyeSight is the only lane-keeping component on Subarus. Some models have radar sensors for blind spot or cross traffic warning, but those aren't used for lane keeping or forward collision prevention.

Re: Tesla in self-drive mode slams into police car in Orange County

#90

It boggles my mind that any self driving system would knowingly drive into any object, no matter what it is, as large as a police car. I'd be willing to bet that Elon's insistence on only using cameras is combining with the flashing lights to effectively make objects that appear in some frames of video, and not in others, as the camera recovers from the flash... and this makes it appear as noise, and below a certain…

AFAIK the latest betas have switched to a cutting edge “transformer” model (or models?). I wonder if it’s prone to hallucinations like its LLM cousins. On a different note, I like how Mercedes changes the color of external lights when in self-driving mode to make others aware. A bit like outsourcing the responsibility externally - but maybe this is the way forward until the tech is reliable/better than people.

transformer and LLM aren't cousins.

it's like saying transistor and cpu is a cousin...

Post reply on HN