Scenarios in which Tesla FSD Beta 9.0 fails • Hacker News Board

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#421

Post by aeternum » Tue, Jul 13, 2021, 3:14 AM UTC

SloopJon wrote: post #86

nikhizzle wrote: post #3

Does anyone have any insight as to why regulators allow a beta like this on the road with drivers who are not trained to specifically respond to it's mistakes? IMHO this puts everybody and their children at risk so Tesla can beta test earlier, but I would love to corrected.

The Twitter thread is lacking in attributions, but I saw most of these some months back after a post about scary FSD behavior. I watched with morbid curiosity, and a rising level of anger. The guy with the white ball cap repeatedly attempted a left turn across two lanes of fast-moving traffic, with a drone providing overhead views. He seemed smart, aware, and even a bit conservative in intervening. Nonetheless, I cou…

Do people sign up to be beta testers whenever a new 15 year-old with a permit gets behind the wheel?

It's even crazier that we allow that. The cars that student drivers learn in typically do not have dual control so it's quite difficult to intervene. Every highschool has a story about how a student slammed the accelerator instead of the brake.

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#422

Post by ahahahahah » Tue, Jul 13, 2021, 3:16 AM UTC

ricardobeat wrote: post #344

Syonyk wrote: post #11

I don't understand how something this broken is allowed to operate on public roads. If I drove like Tesla's FSD seems to based on the videos I've seen, I'd be pulled out of my car and arrested on (well founded) suspicions of "driving while hammered." After a decade of work, it's not capable of doing much beyond "blundering around a city mostly without hitting stuff, but pay attention, because it'll try to hit the mos…

This is a collection of videos from thousands of hours of driving. I’m sure you can do much worse from human drivers…

There's probably barely even 1000 hours on all of youtube of FSD Beta 9.0. It looks like these clips are actually from just 3 videos which in total are less than 1 hour of video.

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#423

Post by aeternum » Tue, Jul 13, 2021, 3:18 AM UTC

paxys wrote: post #316

the8472 wrote: post #201

Earlier quoted context omitted.

I think you misunderstand the argument. It is that if, hypothetically, FSD really did save human lives on average then it should be accepted as the default mode of driving. It would be a net win in human lives after all. But the "should" can also acknowledge that people irrationally won't accept this net life-saving technology because it will redistribute the deaths in ways which they're not accustomed to. So it's as…

I don't agree with the former argument either. I'm not going to accept a self driving system unless it increases my personal safety. If the system doubles my accident rate but cuts that of drunks by a factor of 10 (thus improving the national average), it isn't irrational to not want it for myself.

Why do you believe the system is likely to double your accident rate?

Statistically autopilot gets in fewer accidents than the average human driver. And if you believe you are an above average driver, why do you believe you would be unable to intervene if needed?

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#424

Post by tw04 » Tue, Jul 13, 2021, 3:37 AM UTC

TheParkShark wrote: post #348

raisedbyninjas wrote: post #314

Earlier quoted context omitted.

They currently charge $10k for FSD. Even if they don't drop any of their cameras and processing equipment, $500 LIDAR is a 5% increase. Is it just me or are we politely ignoring Musk's reluctance to admit his bold proclamations against LIDAR are the biggest factor in their design decisions.

Andrej Karpathy already explained they were getting too much noise from Radar. Adding LIDAR would only complicated it. Unless you have better knowledge than Andrej Karpathy, I’ll go with his word on this. Edit: nitrogen I can’t comment directly on yours, so I’ll ass it here. That’s a good goal to have and they tried that for well over two years. They found too much noise and confusion. If Radar says one thing and Vis…

> If Radar says one thing and Vision says another over and over who do you trust?

Neither? If you don't have a tiebreaker you disengage and let the human decide. We solved this problem in systems long ago. You either fail open or fail closed depending on the system. In this case it seems like you'd always fail to the safest option which is alert the human.

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#425

Post by villgax » Tue, Jul 13, 2021, 3:39 AM UTC

This is what is needed to pit against human drivers & other FSD software from manufacturers.

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#426

Post by serverholic » Tue, Jul 13, 2021, 3:40 AM UTC

rasz wrote: post #115

serverholic wrote: post #72

Earlier quoted context omitted.

We are going to need a fundamental breakthrough in AI to achieve the level of FSD that people expect. Like a convolutional neural network level breakthrough.

Most problems in those clips didnt require general AI, they were caused by shit vision algorithms. Car didnt spot huge ass monorail columns ...

I didn't say general AI.

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#427

Post by tw04 » Tue, Jul 13, 2021, 3:43 AM UTC

brianwawok wrote: post #273

BugsJustFindMe wrote: post #262

Earlier quoted context omitted.

The post I was responding to said lidar, not radar. But if you want to switch to radar, we can talk about that too. > camera sees air, radar sees wall, slam on brakes Seeing bridges as walls is not a fundamental property of radar. That's an implementation problem. If cars are doing radar poorly, maybe the fix is to start doing it less poorly instead of throwing it away entirely.

Well, that was specifically about blending two different sensors with different characteristics. For you walking, it would be like blending your eyes with your nose. If your eyes tells you the floor is safe, and your nose smells something bad, do you stop? Anytime you have two different sensors with different characteristics, you want "the best". Your body uses your eyes to see where to walk, and your nose to test if…

> If your eyes tells you the floor is safe, and your nose smells something bad, do you stop?

If my eyes tell me the floor is safe, but my ears hear it creaking and my feet feel it sagging I'm going to stop in my tracks/back up. If Tesla can't handle correlating multiple different types of sensors they aren't ever replacing a human driver.

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#428

Post by londons_explore » Tue, Jul 13, 2021, 3:59 AM UTC

sbierwagen wrote: post #393

TheParkShark wrote: post #348

Earlier quoted context omitted.

Andrej Karpathy already explained they were getting too much noise from Radar. Adding LIDAR would only complicated it. Unless you have better knowledge than Andrej Karpathy, I’ll go with his word on this. Edit: nitrogen I can’t comment directly on yours, so I’ll ass it here. That’s a good goal to have and they tried that for well over two years. They found too much noise and confusion. If Radar says one thing and Vis…

Andrej works for Tesla, so yes, I don't believe him. If he contradicted his boss in public, Musk would just fire him and hire someone who will repeat the party line. I won't believe anything he says about lidar until he's no longer at Tesla, or until they have a car that's not trying to ram concrete pillars.

Andrej has said some stuff in public that certainly implies he's prioritizing the 'party line' on stuff like HD maps and lidar than his academic stats knowledge.

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#429

Post by londons_explore » Tue, Jul 13, 2021, 4:04 AM UTC

BugsJustFindMe wrote: post #186

AlexandrB wrote: post #157

Earlier quoted context omitted.

The monorail posts and planters would be trivially handled by LIDAR. Tesla's aversion to time of flight sensors often strikes me a premature given our level of planning/perception technology.

They should be trivially handled by stereopsis and structure from motion as well. Stereo+time photogrammetry has been solved well enough to not steer directly towards large obstacles for decades. Overreliance on machine learning pixel models to classify everything in view is the real problem.

Stereopsis doesn't work for these columns because the car only has a single camera seeing them.

Structure from motion gets bad results from textureless solids where the edges might be a moving shadow.

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#430

Post by Uehreka » Tue, Jul 13, 2021, 4:19 AM UTC

fastball wrote: post #292

BugsJustFindMe wrote: post #205

Earlier quoted context omitted.

> What are you proposing? If someone is overweight with high blood pressure and a higher risk for a heart attack we should take away their license? This does not seem like an important nit to pick, so I'm wondering why you're picking it. The number of people killed or injured because a driver had a heart attack at the wheel is vanishingly small compared to drunk or distracted driving. So let's reframe your question a…

Clearly it's not that simple, otherwise we'd have done it already.

We… already have? Drunk driving is grounds for license suspension, heavy fines and jail time in (I’m pretty sure) every US state. We may still have drunk drivers, but it seems odd to suggest that we aren’t doing anything about it. Cops, judges and juries take it super seriously. If you get caught drunk driving in America: Do Not Pass Go, Do Not Collect $200.