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

Tesla's self driving algorithm's overlay [video]

tesla.com

Post reply on HN
Print view View original

51–60 of 658 posts

Re: Tesla's self driving algorithm's overlay [video]

#51

Post by Symmetry »

veeralpatel979 wrote: post #29
dylan604 wrote: post #2

I have often wondered about this, so I do find this interesting. Of all of the info presented, one question I have is how does the AI decide when to go at a 4-way stop? In real life, I'm am constantly amazed at how confusing a 4-way stop is to humans. Not that I ever had any doubt into how complicated real-time video analysis could be, this just makes my appreciation of the complexity of the problem that much more qu…

I've wondered why cars don't have a built-in communication mechanism. I should be able to broadcast somehow (using a voice message?) to nearby cars that "I'm going now, don't go".

That works if every car has the same communication mechanism and is honest and nobody interferes and everybody's semantic understanding of the messages is the same.

Re: Tesla's self driving algorithm's overlay [video]

#52

Post by judge2020 »

thedance wrote: post #30

I wonder why they would let the public see this. To the critical eye it looks like it barely works. It loses its mind right at the beginning, around 0:06-0:07, veers to the right for no reason and then rolls through a stop sign without stopping. It spends the rest of the video hunting left and right like a drunk.

Some have speculated that it's a human driving and not autopilot (plus autopilot still has a warning that it's only meant for highways).

Re: Tesla's self driving algorithm's overlay [video]

#53

Post by Groxx »

IceSentry wrote: post #37
thedance wrote: post #30

I wonder why they would let the public see this. To the critical eye it looks like it barely works. It loses its mind right at the beginning, around 0:06-0:07, veers to the right for no reason and then rolls through a stop sign without stopping. It spends the rest of the video hunting left and right like a drunk.

The video is clearly sped up and it looks like it did wait at the stop sign in the first part, it just didn't wait long.

yea, there's a big "stop" icon at the top, and it goes down to 2mph (afaict not zero). a rolling stop in any case.

Re: Tesla's self driving algorithm's overlay [video]

#54

Post by w_t_payne »

veeralpatel979 wrote: post #29
dylan604 wrote: post #2

I have often wondered about this, so I do find this interesting. Of all of the info presented, one question I have is how does the AI decide when to go at a 4-way stop? In real life, I'm am constantly amazed at how confusing a 4-way stop is to humans. Not that I ever had any doubt into how complicated real-time video analysis could be, this just makes my appreciation of the complexity of the problem that much more qu…

I've wondered why cars don't have a built-in communication mechanism. I should be able to broadcast somehow (using a voice message?) to nearby cars that "I'm going now, don't go".

Vehicle-to-X (A generalisation of vehicle-to-vehicle communication) https://en.wikipedia.org/wiki/Vehicle-to-everything

Re: Tesla's self driving algorithm's overlay [video]

#55

Post by winternett »

quantdev wrote: post #23
winternett wrote: post #13

Earlier quoted context omitted.

The system isn't AI based... It's currently based on static programming. This is why a development team would have to think of every possible scenario that could happen on roads to make self driving cars really work. You'd have to ask all kinds of wild questions too like: "What if kids played a prank and re-painted road lanes on a highway at 50MPH?" "What if a sand storm happened?" "What would the car do if a tire bl…

The system is absolutely AI based today. There is no hardcoding like this. Hardware 3 in Tesla cars is an ASIC meant to run neural nets ( https://en.wikipedia.org/wiki/Tesla_Autopilot#Hardware_3 ). Why are you so confidently stating things you don't know about?

From your own cited source:

At this level, the car can act autonomously but requires the driver to be prepared to take control at a moment's notice.[103][104] HW1 is suitable only on limited-access highways, and sometimes will fail to detect lane markings and disengage itself. In urban driving the system will not read traffic signals or obey stop signs. This system also does not detect pedestrians or cyclists,[105] and while AP1 detects motorcycles,[106] there has been two instances of AP rear-ending motorcycles.[107]

https://en.wikipedia.org/wiki/Tesla_Autopilot#Hardware_3

Re: Tesla's self driving algorithm's overlay [video]

#56

Post by jamestimmins »

jedberg wrote: post #42
daenz wrote: post #38

>Vision fps: ~18 Surely that can't be right

Probably is. The computer power necessary to go higher would use all the battery and take up a ton of space. If you've seen other self driving cars, they usually have no trunk because the entire thing is server racks.

Maybe this is ridiculous, but is it feasible that Tesla plans to use the SpaceX Starlink satellites to offload data for processing remotely?

Re: Tesla's self driving algorithm's overlay [video]

#57

Post by mitchellgoffpc »

Erlich_Bachman wrote: post #33
jfengel wrote: post #25

Earlier quoted context omitted.

The "vision fps" runs around 12 to 20, so yeah, that's gonna seem kinda jumpy. The display fps is slightly higher, which contributes to a juddering feeling.

I think the more interesting/dangerous jumping is when the perception engine fails to consistently classify/find objects in the scene, between adjacent frames, not the specific fps of the process. This video seems to display some of such inter-frame jumpiness.

Inter-frame jumpy-ness can be smoothed out slightly by the control algorithms, on the assumption of object permanence, as long as the underlying vision NN doesn’t miss things too often.

Re: Tesla's self driving algorithm's overlay [video]

#58

Post by thedance »

baddox wrote: post #40
winternett wrote: post #13

Earlier quoted context omitted.

The system isn't AI based... It's currently based on static programming. This is why a development team would have to think of every possible scenario that could happen on roads to make self driving cars really work. You'd have to ask all kinds of wild questions too like: "What if kids played a prank and re-painted road lanes on a highway at 50MPH?" "What if a sand storm happened?" "What would the car do if a tire bl…

> The system isn't AI based... It's currently based on static programming. This is why a development team would have to think of every possible scenario that could happen on roads to make self driving cars really work. I'm not really sure what this dichotomy is that you're describing. It certainly looks like they're using computer vision to detect the bounding boxes of cars, and the field of computer vision is widely…

> What do humans do when they suddenly have nearly zero visibility on a road?

They cause gigantic pile-ups, because humans are unbelievably bad at driving.

https://abcnews.go.com/US/50-people-injured-69-vehicle-pileu...

Re: Tesla's self driving algorithm's overlay [video]

#59

Post by anchpop »

Traster wrote: post #21
ArtWomb wrote: post #7

I believe this was released as part of Tesla's Autopilot hiring announcement. More details here: https://www.tesla.com/autopilotAI Definitely someone's dream job ;) I particularly like the applicant query: "Tell us, what extraordinary work you have done?"

Two issues, first: I thought fully autonomous driving was meant to be done by now? Second, don't Tesla & SpaceX have a reputation for being a terrible place to work, with Musk expecting everyone to be working as hard (or harder) than him, and firing people in weird and capricious ways.

> I thought fully autonomous driving was meant to be done by now?

It was, but software often goes over-schedule, especially in a brand new field.

Re: Tesla's self driving algorithm's overlay [video]

#60

Post by IceSentry »

Traster wrote: post #26
yarg wrote: post #18

Interesting, but is it just me or does it seem overly jumpy?

One of the real challenges of these systems is that even a very good system only needs to lose confidence in some object for a very short period of time for the system to lose confidence in its navigation and return control to the user. The easy way to get around that issue is to categorize objects and then just ignore them out of the context you expect them to appear in -- which is how Tesla managed to kill a pedest…

That event was done by Uber and if the human backup driver was paying attention, as they were supposed to because it was a test run, it wouldn't have happen. Yes the software was at fault of doing what you are saying, but the incident was a test run and the driver should have caught it and report to the engineer, not play on their phone and not pay attention.
Post reply on HN