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

Tesla influencers tried coast-to-coast self-driving, hit debris before 60 miles

electrek.co

Post reply on HN
Print view View original

351–360 of 410 posts

Re: Tesla influencers tried coast-to-coast self-driving, hit debris before 60 miles

#351

Post by jjav »

rob74 wrote: post #43
agildehaus wrote: post #30

A friend of mine who loves Tesla watched this video and said "many humans would have hit that". I feel we'll be hearing a lot of that excuse.

Well, I have to admit that your friend has a point. Humans are bad at reacting quickly and correctly to unexpected situations, and some debris large enough to damage your car showing up from out of nowhere after several hours of boring driving along a largely straight highway with little traffic is definitely one of these situations. But a self-driving system worth its salt should always be alert, scanning the road a…

> Well, I have to admit that your friend has a point. Humans are bad at reacting quickly and correctly to unexpected situations, and some debris large enough to damage your car showing up from out of nowhere

I read this comment before seeing the video and thought maybe the debris flies in with the wind and falls on the road a second before impact, or something like that.

But no, here we have bright daylight, perfect visibility, the debris is sitting there on the road visible from very far away, the person in the car doing commentary sees it with plenty of time to leisurely avoid it (had he been driving).

Nothing unexpected showed up out of nowhere, it was sitting right there all along. No quick reaction needed, there was plenty of time to switch lanes. And yet Tesla managed to hit it, against all odds! Wow.

My impression of Tesla's self driving is not very high, but this shows it's actually far worse than I thought.

Re: Tesla influencers tried coast-to-coast self-driving, hit debris before 60 miles

#352

Post by xnx »

ModernMech wrote: post #248
newyankee wrote: post #206

Earlier quoted context omitted.

I feel Waymo's LIDAR might have worked here

100% it would have. One of the main things the LiDAR system does is establish a "ground plane", which is the surface on which the car is expected to drive. Any hole or protrusions in that plane stick out like a sore thumb to a LiDAR system, you'll be able to see it in the raw data without much of a feature detector, so detecting them and reacting is fast and reliable. Contrast with Tesla's "vision-only" system, which…

In addition to detecting the object, Waymo has to make some determination about the material. Rigid heavy metal = slam on the brakes and/or swerve. Piece of tire or plastic bag = OK to run over if swerving or hitting the brakes would be more dangerous. Really hard problem that they're concerned about getting right before they open up highway driving.

Re: Tesla influencers tried coast-to-coast self-driving, hit debris before 60 miles

#353

Post by jjav »

foobarian wrote: post #253
ModernMech wrote: post #245

Earlier quoted context omitted.

If you don't disagree on the characterization of the technical details, then you must realize how very fair it is for us to criticize the system for failing in the exact way it's predicted to fail. We don't need 1000 more trials to know that the system is technically flawed.

What if there is no debris the other 999 times, and the system works fine? The video does not give me that information as a prospective Tesla customer. This looks like a fluke to me.

> What if there is no debris the other 999 times, and the system works fine?

This argument makes no sense. I take it that you're saying that if we provide the Tesla a road which contains nothing to hit, it won't hit anything?

Well, sure. Also, not interesting.

In a real world drive of almost 3000 miles there will nearly always be things to avoid on the way.

Re: Tesla influencers tried coast-to-coast self-driving, hit debris before 60 miles

#354

Post by jjav »

fuzzfactor wrote: post #307
mynameisvlad wrote: post #261

Earlier quoted context omitted.

Those 999 other times, the system might work fine for the first 60 miles . This is a cross-country trip. LA to New York is 2776 miles without charging. It crashed the first time in the first 2% of the journey. And not a small intervention or accident either. How you could possibly see this as anything other than FSD being a total failure is beyond me.

>asking a different question: how good would the FSD system be at completing a coast-to-coast trip? >They made it about 2.5% of the planned trip on Tesla FSD v13.9 before crashing the vehicle. This really does need to be considered preliminary data based on only one trial. And so far that's 2.5% as good as you would need to make it one way, one time . Or 1.25% as good as you need to make it there & back. People will…

> This really does need to be considered preliminary data based on only one trial.

Statistically yes, but look at the actual facts of the case.

A large object on the road, not moving, perfect visibility. And the Tesla drives straight into it.

Not hitting static objects in perfect visibility is pretty much baseline requirement #1 of self driving. And Tesla fails to meet even this.

Re: Tesla influencers tried coast-to-coast self-driving, hit debris before 60 miles

#355

Post by jjav »

dmix wrote: post #246
rkomorn wrote: post #224

Earlier quoted context omitted.

Are you saying "they let the car drive on its own but were still paying attention"? Not sure how to take this reply.

That a human was still in the loop in addition to a computer and both missed it.

> That a human was still in the loop in addition to a computer and both missed it.

Listen to the audio in the video. The humans do see it and talk about it for a long time before the car hits it. Had a human been driving, plenty of time to avoid it without any rush.

They do nothing to avoid it presumably because the whole point of the experiment was to let the car drive, so they let it drive to see what happens. Turns out Tesla can't see large static objects in clear daylight, so it drives straight into it.

Re: Tesla influencers tried coast-to-coast self-driving, hit debris before 60 miles

#356

Post by ryandrake »

anonymars wrote: post #339
whatevaa wrote: post #324

Earlier quoted context omitted.

Why did you buy FSD if you are in Europe?

I will never understand why people ask why those who bought a thing expected to get that thing Isn't that the bare minimum requirement for how commerce is supposed to function?

A surprising chunk of HN commenters earnestly believe that "caveat emptor" should be some kind of horrible default way of life. Like businesses should be able to sell anything they want, without regulation, as long as they can convince people to buy those things. And if those things don't work--well it's 100% the customer's fault for being so gullible and not being responsible for the company's quality.

Re: Tesla influencers tried coast-to-coast self-driving, hit debris before 60 miles

#357

Post by kgwgk »

richrichardsson wrote: post #229
kgwgk wrote: post #210

Earlier quoted context omitted.

It’s not like you would expect any inconvenience from being rear-ended or anything so why would you care?

The point is "If you *have* to hit the brakes to avoid a collision in front of you". Obviously if you have a tailgating idiot behind and you can avoid the collision in front without forcing them to rear-end you, then you do that; but that's not always possible, and in that case the blame is then entirely on the person who hit you from the rear, and *not* on you for rear-ending someone.

> Obviously if you have a tailgating idiot behind and you can avoid the collision in front without forcing them to rear-end you

That kind of consideration is where having traffic tailgating factors into it. A risk of collision (or a minor collision) may be a better option than a certain (or worse) collision with the tailgating idiot.

Re: Tesla influencers tried coast-to-coast self-driving, hit debris before 60 miles

#358

Post by burnt-resistor »

cosmicgadget wrote: post #283
raisedbyninjas wrote: post #127

Earlier quoted context omitted.

Sensor fusion is too hard is the new cope.

To paraphrase sarcastically: "But what happens when the sensors disagree???"

[Insert red herring about MCAS and cop out about how redundancy is "hard" and "bad" "complexity".]

Have a minimum quorum of sensors, disable one if it generates impossible values (while deciding carefully what is and isn't possible), use sensors that are much more durable, reliable, and can be self-tested, and integration and subsystem test test test thoroughly some more.

Re: Tesla influencers tried coast-to-coast self-driving, hit debris before 60 miles

#359

Post by jjav »

SillyUsername wrote: post #60

A lot of apologists say that "a human would have hit that". That's kind of irrelevant, this technology is meant to be safer and held to a higher standard. Comparing to a human is not a valid excuse...

> That's kind of irrelevant, this technology is meant to be safer and held to a higher standard.

True, but not even relevant to this specific example. Since the humans clearly saw it and would not have hit it, so we have a very clear example where Tesla is far inferior to humans.

Re: Tesla influencers tried coast-to-coast self-driving, hit debris before 60 miles

#360

Post by bambax »

spacebanana7 wrote: post #76
bambax wrote: post #57

What is the point of "testing" something that everyone knows doesn't work and only exists because a serial liar said so? If Musk says you will survive a fall from a 90-stories high-rise... don't test it.

We should rigorously test the things we’re most skeptical of. You shouldn’t take pills a stranger gives you at a music festival without checking them for loads of things, for example. Even if you don’t ever intend on consuming them it’s nice to have specific accusations with evidence.

Or you can discard them right away. I'm not the police, I don't investigate strangers.
Post reply on HN