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

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#361

Post by TheParkShark » Mon, Jul 12, 2021, 11:11 PM UTC

cromwellian wrote: post #355

TheParkShark wrote: post #335

Earlier quoted context omitted.

I see this argument a lot. This is a small beta test and I have yet to see anything better on a larger scale. Most drivers shouldn’t be allowed on public roads yet here we are. Why are you so full of fear and anger over this? Isn’t this a place for open discussion?

Because every year, Tesla keeps saying it's going to ship soon. It's clear this isn't even close to ready, I'd say it's 10 years from shipping, if not back to the drawing board. Tesla is having people pay for a self driving package as an option, and years later, it still isn't available. Shouldn't they refund their owners? I mean, just look at the visual representation of what the FSD is showing. It has no temporal s…

The visual representation isn’t an accurate measurement of what the car is actually seeing, I think Tesla cleaning this up (which is has on the newest FSD) will be a clear way to show people what it sees. I’d like to see Tesla do more to calm doubts about this.

If you watch the full videos and not a mash up, you’ll see most drivers drive long routes without any problems with the software. It’s the edge cases you’re seeing and yes you can make a compilation with them and make it look horrible.

Honest question, do you think a Tesla that doesn’t recognize whats in front of it to 100% should stop in the road? I saw this recently with Waymo and it stayed blocking traffic until a Waymo tech arrived then the car tried to drive away from the tech. This doesn’t seem like the better answer to me, but I don’t have one.

As far as comparing it to a teenage driver, they are the most dangerous drivers on the road and Tesla’s happen to be the safest rated cars in accidents. Interesting combo.

Can you please link an article about FSD and deaths? There have been a few on auto pilot when the drivers are not paying attention but I’m not aware of a single FSD death.

Why can’t other companies make self driving software/hardware that a normal person can afford? All other options are insanely expensive and out of reach of anyone but millionaires.

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#362

Post by AtlasBarfed » Mon, Jul 12, 2021, 11:12 PM UTC

Syonyk wrote: post #198

AtlasBarfed wrote: post #172

Earlier quoted context omitted.

RE: determining what humans did was right to take over It's a QA department. If there is a failure hot spot, then take a bunch of known "good" QA drivers through that area. Assign strong weight to their performance/route/etc. It's interesting reading through all this, I can see a review procedure checklist: - show me how you take hotspot information into account - show me how your QA department helps direct the softw…

> Also, construction absolutely needs to evolve convergently with self driving. Cones are... ok, but some of those people leaning on shovels need to update systems with information on what is being worked on and what is cordoned off. No. If the car cannot handle random obstructions and diversions without external data, it cannot be allowed on the road. Construction is often enough planned ahead of time, but crashes h…

Then broadcast an alert on a status channel that the AI can take into account. They're going to do it anyway for traffic and other hazards.

Doing so ahead of time for planned construction is not a big ask.

"And if you can handle that on-car, which I consider a hard requirement, then why not simply use that logic for all the cases involving detours and lane closures?"

You're making the same mistake Musk made when insisting that the car be able to navigate regardless of location or connectivity. Ignoring the ability of networking/radio broadcast/internet databases to provide vastly more deep information pools is a big mistake.

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#363

Post by qeternity » Mon, Jul 12, 2021, 11:12 PM 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…

Karpathy is the guy responsible for the results we see in these videos. Why would you trust his word about any of this?

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#364

Post by nitrogen » Mon, Jul 12, 2021, 11:17 PM UTC

schmorptron wrote: post #285

jazzyjackson wrote: post #209

Earlier quoted context omitted.

I feel like the Xbox Kinect from 2010 would be a better vision solution than what they’ve got here, at least for the 20 feet ahead of you.

Until you have multiple cars trying to project multiple arrays of infrared dots onto objects 20 meters away in bright sunlight and then getting an accurate reading on which of the array kerfuffle is theirs.

If I remember right, the Kinect actually still handled this reasonably well for a couple of overlapping sensors. In a driving scenario other cars' lightfields should also be blurred.

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#365

Post by meepmorp » Mon, Jul 12, 2021, 11:19 PM UTC

CyanLite2 wrote: post #317

potatolicious wrote: post #213

Earlier quoted context omitted.

Yep, also over-promising that legacy cars with the "FSD-readiness" upgrade would be guaranteed to support FSD was a completely unforced error. At this point it's pretty obvious none of the original "FSD ready" cars Tesla shipped will be able to handle L4/L5 autonomy hardware-wise. Arguably none of the cars currently rolling off the line will be either - others have made the point in the thread: while nobody has reall…

I think there's a clear space in North America for a viable L3 system. That is, completely hands-free driving at slow speeds (30-50mph) that allows me to eat a sandwich, respond to a few emails, and text on my phone without having to look at the road or grab the wheel every 15 seconds. Sure, ping me if I'm coming up on a construction zone or if it pours down raining. And don't worry about making unprotected lefts and…

I think that sounds good on paper, but the reality would likely be pretty lousy. Humans are still the critical safety component in the system, and all we've done is made it easier for them to have no idea what's going on in a situation before they need to respond to a crisis.

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#366

Post by TheParkShark » Mon, Jul 12, 2021, 11:26 PM UTC

ahahahahah wrote: post #357

TheParkShark wrote: post #342

Earlier quoted context omitted.

Please share them if you have. As someone in another thread said: Tesla has the most advanced self driving software currently in development in the world. If there’s something better you can feel free to share that with your other videos.

> Tesla has the most advanced self driving software currently in development in the world This is just nonsense. These videos show them at a disengagement every couple miles. Waymo was at that level 10 years ago and is 10000s of times better now.

Waymo only works in a 50 sq mi area [1] and has it’s own issues [2]. So no, this is not nonsense.

[1] https://arstechnica.com/cars/2021/05/why-hasnt-waymo-expande... [2] https://www.bbc.com/news/technology-57144755

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#367

Post by BrissyCoder » Mon, Jul 12, 2021, 11:26 PM UTC

moralestapia wrote: post #93

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…

Because Elon doesn't operate under the same jurisdiction as us common people. Try calling out a random diver a 'pedo', or perform the most cynical kind of market manipulation (then laughing to the SEC at their face) and your outcome will be very different. It's Animal Farm all over.

It's been a while since I read it... What's the Animal Farm reference here?

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#368

Post by pie420 » Mon, Jul 12, 2021, 11:33 PM UTC

heavyset_go wrote: post #194

moralestapia wrote: post #93

Earlier quoted context omitted.

Because Elon doesn't operate under the same jurisdiction as us common people. Try calling out a random diver a 'pedo', or perform the most cynical kind of market manipulation (then laughing to the SEC at their face) and your outcome will be very different. It's Animal Farm all over.

> Try calling out a random diver a 'pedo', or perform the most cynical kind of market manipulation (then laughing to the SEC at their face) and your outcome will be very different. Not just any random diver, a diver who had just helped rescue the lives of several children and adults from what was an internationally known emergency. If any of us had said what Musk said about the diver, we'd be rightfully dragged throu…

Are you actually implying that Elongated Muskrat wasn't dragged through the mud for that? The fact that everyone still talks about it years later should be proof enough, but keep living in your little world buddy

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#369

Post by sudosysgen » Mon, Jul 12, 2021, 11:41 PM UTC

maverick-iceman wrote: post #242

moralestapia wrote: post #148

Earlier quoted context omitted.

Step 1: Become the CEO/major shareholder of a billion+ dollar company. Step 2: Publicly lie about some purported acquisition with the purpose of manipulating the stock of said company. We can talk after you've done that, from jail probably.

I think the SEC gives insiders and C-suite some kind of free pass as long as they don't time their 'overtlyoptimistic' takes on their company with their stock sales or vest. The SEC is giving Musk free reign because they know that he can't leave the company because the two entities are so intertwined. Musk wealth is effectively just paper wealth. The SEC doesn't care that investment bankers consider it real enough to…

My memory might be falling me, but wasn't Elon Musk's manipulation actually timed to coincide with his performance bonus, and directly ended with millions of dollars leaving Tesla coffers and endint up in Elon's pocket?

Re: Scenarios in which Tesla FSD Beta 9.0 fails

#370

Post by patentatt » Mon, Jul 12, 2021, 11:42 PM UTC

BugsJustFindMe wrote: post #284

brianwawok wrote: post #273

Earlier quoted context omitted.

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…

Unfortunately your example is inapt because the center of your vision and your peripheral vision may as well be entirely separate systems that don't overlap, and the way brains apply focus doesn't translate to how cameras work. Your scenario is closer to asking about the center of your focus saying the path in front of you is clear and your peripheral vision detecting inbound motion from the side. Peripheral motion d…

This. When sensors disagree, you need to proceed with caution by taking a defensive stance and seek out more information. If a self-driving car can’t do that, then it can’t self drive really. I’m bearish on self driving, my uninformed take is that it is something that will require very near general AI to pull off acceptably. So it’s kind of silly, again in my totally uninformed opinion, to focus on using our current methods to do self driving. The focus needs to be on pushing toward general AI first, then just ask it to drive.