Video of Tesla FSD almost hitting pedestrian receives DMCA takedown • Hacker News Board

Re: Video of Tesla FSD almost hitting pedestrian receives DMCA takedown

#301

Post by Matthias247 » Fri, Sep 17, 2021, 5:49 PM UTC

xxs wrote: post #204

darkmarmot wrote: post #154

Earlier quoted context omitted.

I recently avoided a startup prospect because they were looking to build a car OS that wasn't hard real-time. The very idea that they're trying to develop such things that might intermittently pause to do some garbage collection is freaking terrifying.

What has hard real time have to do with garbage collection? You can have concurrent GCs (with [sub]millisecond pauses, or no Stop-the-world at all) but you also need 'hard real time' OS to begin with. Heck, even opening files is far from real-time. Then you need: not-blocking data strcutures, not just lock-free - that are much easier to develop. Pretty much you need forward guarantees on any data structure you'd use.

You usually need garbage collection because you are allocating in the first place. And allocating and releasing adds some non-determinism. You apparently don't know how much exactly needs to be allocated - otherwise you wouldn't have opted for the allocations and GC. That non-determinism translates to a non-determinism in CPU load, as well as to "I'm not sure whether my program can fulfill the necessary timing constrains anymore".

So I kind of would agree that producing any garbage during runtime makes it much harder to prove that a program can fulfill hard realtime guarantees.

Re: Video of Tesla FSD almost hitting pedestrian receives DMCA takedown

#302

Post by kzrdude » Fri, Sep 17, 2021, 5:50 PM UTC

koonsolo wrote: post #131

Maybe it's just me, but it seems the car wanted to make a right turn on the right part of that side street. There are no pedestrians there. It's actually the interference of the driver that stops the turn halfway, and steers towards the people at the left starting to cross that street. I wonder what would have happened when the driver didn't interfere. I guess that the right turn would have been completed without any…

There are a few problems:

Erratic navigation and driving. 1 second earlier the car was heading straight. The car placement was firmly to the left edge of its lane. A safe driver would slow down towards the turn and use the right hand side of the road - pedestrians will recognize this "body language" of the car and understand the situation better.

In our peripheral vision, the car placement and "body language" is probably more important than ostensible signals such as the turn signal (!)

This kind of "body language" is something we learn when driving and automatic drivers should adhere to it too, so that they can be predictable and interpretable drivers.

Re: Video of Tesla FSD almost hitting pedestrian receives DMCA takedown

#303

Post by tsimionescu » Fri, Sep 17, 2021, 5:52 PM UTC

antattack wrote: post #155

staticassertion wrote: post #54

Earlier quoted context omitted.

The car veers pretty aggressively and suddenly. "Almost hitting" or not, this looks dangerous. > This sort of close interplay between pedestrians and cars is very common in cities. Yeah and I'm usually yelling at the idiot that wasn't looking where they were driving.

In this case pedestrians waved back to the driver who apologized as they were far away. If they were closer car would have probably stopped as AP stops and slows down near pedestrians.

When it's not driving straight into them, that is. Which it has done quite a few times.

Re: Video of Tesla FSD almost hitting pedestrian receives DMCA takedown

#304

Post by Lendal » Fri, Sep 17, 2021, 5:54 PM UTC

ChicagoBoy11 wrote: post #142

ChrisKnott wrote: post #63

Maximise the video and watch the left hand HUD pane from 0:10 to 0:11. The dotted black line coming from the front of the car (which I am assuming is the intended route) quickly snaps from straight ahead, to a hair pin right, to a normal right turn. Ignoring the fact that the right turn happened to be into a pedestrian crossing with people on it - what was the car even trying to do? The sat-nav shows it should have j…

I've made this argument here a few times and am always shot down, but I think its important to highlight that the airline industry has an extremely robust history of automation and HCI in critical transportation scenarios and it seems to me that all the lessons that we have learned have been chucked out the window with self-driving cars. Being able to effectively reason about what the automation is doing is such an i…

The mechanism by which those lessons were learned involved many years full of tragedy and many fatalities including many famous celebrities dying in those plane crashes. Obviously, we do not want to follow that same path, but at the moment that's exactly the path we're on.

The US govt isn't going to do anything until there's a public outcry, and historically there won't be a public outcry until there's a bunch of famous victims to point to.

Re: Video of Tesla FSD almost hitting pedestrian receives DMCA takedown

#305

Post by dundarious » Fri, Sep 17, 2021, 5:54 PM UTC

oneplane wrote: post #276

dundarious wrote: post #233

Earlier quoted context omitted.

Tesla’s “FSD” is only a little bit better than drunk drivers, whom we punish severely whenever caught, even before any accident occurs. The fact that enforcement is patchy is irrelevant — drunk driving is deemed serious enough to be an automatic infraction. Also, most of the time drunk drivers are not actually that bad at moment-to-moment driving. That’s why almost everyone worldwide used to do it! You can still do t…

This assumes FSD-to-drunk-driver analogy means FSD has to be a drunk driver (or a student driver as commented elsewhere) all the time, so always making the bad judgement and slow reaction like a drunk driver would. I think that some form of responsibility has to be assigned to the FSD in some way (the manufacturer? the user? some other entity or a combination?) regardless but I haven't found any clear case of how tha…

I think you’ve missed my point, while adding some additional information.

- drunk drivers are also not uniformly awful drivers: they can drive OK for the most part

- they still drive unacceptably poorly

- we strictly punish them on detection, before any potential accident

- drunk drivers and FSD have more in common with each other than competent drivers and FSD

- why is FSD not held to such a preventative standard?

One can argue that FSD is like a drunk driver driving a student training car with two sets of pedals and two steering wheels, and the Tesla owner/driver is like a driving instructor. But driving instructors are trained and paid to be quite vigilant at all times. Tesla play a sleight of hand and say it’s a labor saving technology, but also you need to be able to behave like a trained and paid driving instructor... that is a conspicuous contradiction.

And I’m ignoring future FSD capabilities because while I’d be happy for it to come about, we should discuss the present situation first, and I don’t believe it’s a good example where sacrificing lives now is acceptable in order to potentially save lives in the future.

Re: Video of Tesla FSD almost hitting pedestrian receives DMCA takedown

#306

Post by no_butterscotch » Fri, Sep 17, 2021, 5:54 PM UTC

joshribakoff wrote: post #70

I have FSD but not the beta. When cars cut across my lane autopilot will not react for a solid second or so, then proceeds to slam on the brakes after the car is clear. It does not instill confidence to say the least. From my POV it feels like someone put a debounce on the reaction to prevent phantom breaking, to the point the car doesn’t brake when it needs to.

I don't have FSD, but autosteer + cruise-control in my Tesla does this as well.

Additionally I've experienced cases of "phantom braking" (https://www.cbsnews.com/news/automatic-emergency-braking-in-...) which doesn't instill confidence in those features.

Re: Video of Tesla FSD almost hitting pedestrian receives DMCA takedown

#307

Post by rootusrootus » Fri, Sep 17, 2021, 5:55 PM UTC

antattack wrote: post #248

sleepybrett wrote: post #230

Earlier quoted context omitted.

it's sold as an 'autopilot' that means 'automatic pilot' that does not imply hybridization, that implies full automation.

I don't know where you get the idea. Even plane autopilot will not land the plane w/o Pilot's supervision. People who don't read manuals put themselves at risk. Even simple systems like Cruise Control carry plenty of warnings that one should be familiar with when operating (from Ford's manual): WARNING: Always pay close attention to changing road conditions when using adaptive cruise control. The system does not repl…

So pilots will get the distinction, but non-pilots think it means self-driving.

Re: Video of Tesla FSD almost hitting pedestrian receives DMCA takedown

#308

Post by Someone » Fri, Sep 17, 2021, 5:55 PM UTC

ChicagoBoy11 wrote: post #142

ChrisKnott wrote: post #63

Maximise the video and watch the left hand HUD pane from 0:10 to 0:11. The dotted black line coming from the front of the car (which I am assuming is the intended route) quickly snaps from straight ahead, to a hair pin right, to a normal right turn. Ignoring the fact that the right turn happened to be into a pedestrian crossing with people on it - what was the car even trying to do? The sat-nav shows it should have j…

I've made this argument here a few times and am always shot down, but I think its important to highlight that the airline industry has an extremely robust history of automation and HCI in critical transportation scenarios and it seems to me that all the lessons that we have learned have been chucked out the window with self-driving cars. Being able to effectively reason about what the automation is doing is such an i…

> it seems to me that all the lessons that we have learned have been chucked out the window with self-driving cars.

I think it’s unfair to lump all self driving car manufacturers together.

The traditional car companies have been doing research for decades (see for example https://en.wikipedia.org/wiki/VaMP), but only slowly brought self-driving features to the market with part of the slowdown because they are aware of the human factors involved. That’s why there’s decades of research on ways to keep drivers paying attention and/or detecting that they don’t.

“Move fast and break things” isn’t their way of working.

Re: Video of Tesla FSD almost hitting pedestrian receives DMCA takedown

#309

Post by koonsolo » Fri, Sep 17, 2021, 5:56 PM UTC

TheSpiceIsLife wrote: post #289

koonsolo wrote: post #131

Maybe it's just me, but it seems the car wanted to make a right turn on the right part of that side street. There are no pedestrians there. It's actually the interference of the driver that stops the turn halfway, and steers towards the people at the left starting to cross that street. I wonder what would have happened when the driver didn't interfere. I guess that the right turn would have been completed without any…

I don’t know what where you live, but here in the civilised world you’re not legally supposed to turn on to a road when there are freaking pedestrians already crossing it . You’re supposed to wait for them to fully complete their crossing, and yes: the overwhelming majority of drivers here abide by that requirement.

The car didn't cross the zebra path, so it still had time to stop and be compliant.

I drove in US, and I know how slow and relax the traffic is. Come drive in Brussels or any of the big European cities (like Paris) and see for yourself.

Good to know US is civilized and EU is not. The difference is that you just have way more space than us.

Re: Video of Tesla FSD almost hitting pedestrian receives DMCA takedown

#310

Post by zaptrem » Fri, Sep 17, 2021, 5:57 PM UTC

mumblemumble wrote: post #162

joshribakoff wrote: post #70

I have FSD but not the beta. When cars cut across my lane autopilot will not react for a solid second or so, then proceeds to slam on the brakes after the car is clear. It does not instill confidence to say the least. From my POV it feels like someone put a debounce on the reaction to prevent phantom breaking, to the point the car doesn’t brake when it needs to.

I worry a little bit about things like this. If they were just using debouncing on a band-aid rather than dealing directly with the bad input that led to the phantom breaking, then that would seem to imply that the project is falling into the classic death march project trap. It turns out that the last 10% of the project is going to be 90% of the work. Meaning that the original schedule was off by an order of magnitu…

Production AP is a completely different software stack than FSD beta. I think the only things that carry over are the car detection, VRU, and traffic light CV NNs.