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

New video of Tesla crash demonstrates the problem of semi-automated driving

theautopian.com

Post reply on HN
Print view View original

621–630 of 714 posts

Re: New video of Tesla crash demonstrates the problem of semi-automated driving

#621

Post by tannedNerd »

matthewdgreen wrote: post #605
jackmott wrote: post #522

Earlier quoted context omitted.

I have a Tesla, when using just cruise control it has slammed the brakes: because of shadows that confused it because of concern about cars crossing the intersection ahead where there is only the need to gently slow because of a car crossing an intersection ahead that has already crossed! because of a steep bridge that confused it we can’t use cruise control in the car basically, we decided it is too dangerous at wor…

The worst part of phantom braking is not just the need for a sudden human intervention. It’s that the required human intervention is extremely non-intuitive. Intuitively when a car suddenly slows down it feels like the right response is to press on the accelerator: however in my experience this puts you in an unstable regime where you are “fighting” with the car. That is, autopilot continues to try to stop but tempor…

It is interesting that in all the talks with my friends that have a Tesla and have tried other driver assist features from other companies. Is that Tesla is only one we are worried about phantom braking aka a false positive and as such we keep the foot near the gas instead of the brakes.

I’ve also developed a reaction to reach for the stalk with my other hand and disable it at the same time so I’m no longer fighting it.

Re: New video of Tesla crash demonstrates the problem of semi-automated driving

#622

Post by amelius »

BuckyBeaver wrote: post #434
cm2187 wrote: post #407

Earlier quoted context omitted.

If you crash into the car in front of you in your lane, the responsibility is yours both morally and legally. The only car which has a valid excuse is the car immediately following the tesla since the tesla inserted itself into its lane then immediately stopped, not allowing it to build a safety distance.

The video does indeed show that this happened. The driver's account, as quoted here, doesn't even make sense. In the first sentence it says he was driving along in lane 1, the car slowed, and he "felt an impact." The next part is contradictory, citing a different lane (2), a different cruising speed, and then a lane change... and no impact. WTH?

Are you surprised you can be guilty of an accident when you have not hit anyone?

Re: New video of Tesla crash demonstrates the problem of semi-automated driving

#623

Post by jjav »

raphman wrote: post #459
Loic wrote: post #318

Earlier quoted context omitted.

In Germany you would be covered because line change and break, without a reason, is a deliberate dangerous action. This means that the one doing it is responsible of the following crash. For the multiple cars crash, if not clear cut, without video recording, all the insurances pool themselves together and consider the drivers as not responsible. Nobody cannot tell if you stopped, got hit in the back and then the fron…

Huh? In my (German) experience, responsibility for the crash (and thus for the sentence/fine/payout to each driver) is rarely assigned to a single driver in such a case. Instead, it is distributed among participants with e.g. the Tesla driver being 80% at fault for the second driver's damages and the others ramming the second car for 20% (or something like this). I admittedly find it a little bit confusing how many p…

> I admittedly find it a little bit confusing how many people here assign complete blame to exactly one participant.

I'd say there is only one car which did a clearly illegal action (sudden lane change followed by brake checking), so the guilt is 100% on them.

It's also true that the car behind probably could've been more assertive in braking harder to avoid reaching the Tesla (I wish there was a dashcam to see it from a better angle). But they didn't do anything illegal per se.

Also, that second car driver might've been worried of braking too hard to avoid getting themselves rear ended, which is a legitimate concern when having to break hard in traffic. Of course in the end they got rear ended anyway, but couldn't have known that prior to it happening.

Re: New video of Tesla crash demonstrates the problem of semi-automated driving

#624

Post by gfodor »

ncallaway wrote: post #313
modeless wrote: post #197

Earlier quoted context omitted.

Automatic switching between FSD and Autopilot is not any kind of UX problem. There is no reason the driver needs to be aware of it. The driver does not need to take any action at all. The only difference is the type of road. But which software was in use is relevant here because people are trying to blame this on FSD. Whenever you're on a freeway, Autopilot is used automatically instead of FSD, so FSD had nothing to…

> There is no reason the driver needs to be aware of it. If the person operating the vehicle doesn't need to be aware of the difference, then the general public shouldn't need to either when analyzing an accident. It seems baffling to me that we could imagine a scenario where it's less important for the driver of a vehicle to be aware of the difference, than it is for the public when reacting to a news article. I don…

It matters not to the driver or to the general public but to audiences like this one where knowing if a 10+ year old mature system not based on ML did this or a beta, in progress neural network based system did this makes a difference in thinking through the implications of such an error.

Re: New video of Tesla crash demonstrates the problem of semi-automated driving

#625

Post by jjav »

smileysteve wrote: post #580
userbinator wrote: post #363

Earlier quoted context omitted.

No, those are not "sudden stop" situations either. In fact the worst thing to do is to stop suddenly. Even if a wheel falls off, you do not slam on the brakes unless you want to lose even more control.

If a wheel falls off, the driver doesn't have to apply the brakes... A quarter of the car is digging in and melding itself to the asphalt. If the differential fails catastrophically, it applies more braking force than the brakes do, so you'll have brief period of being airborne and then at least 1 rear wheel will sheer off (seen it). If the driveshaft or trailing arms fail the wrong way, they result in a steel shaft…

> If a wheel falls off, the driver doesn't have to apply the brakes... A quarter of the car is digging in and melding itself to the asphalt.

That's not necessarily true. I've been in a car where the rear wheel parted ways with the car and passed us. For a few seconds we wondered where that wheel came from until realizing it was from our car. The car was balanced just fine on three wheels, there was plenty time to lift off the gas and make way to the breakdown lane safely.

Even if it's a more heavily loaded wheel (e.g. front wheel on front engine front wheel drive car), it'll slide on the brake disc or the disk cover (whichever is lower, varies by car). A good amount of sparking but you'll have enough steerage to pull over.

Re: New video of Tesla crash demonstrates the problem of semi-automated driving

#626

Post by ncr100 »

joenathanone wrote: post #604
maury91 wrote: post #492

I'm trying to reason "why the Tesla did stop?", in the second video on TheIntercept website, I can notice a left exit before the point where the Tesla stopped. Something I noticed with Google maps ( I don't have a Tesla, and maybe the navigation system of the Tesla is similar ), is that sometimes thinks tunnels can access the road above them ( in a tunnel near my home Google maps always tells me "turn right" in the m…

I have an X and a 3, this is exactly how the car behaves when it has missed a turn because of a mismatch of map/vision data.

To confirm, when the navigation system fails, the entire call halts?

That seems undesirable, since it would lead to rear-ending, in situations like this.

Re: New video of Tesla crash demonstrates the problem of semi-automated driving

#627

Post by jjav »

jejeyyy77 wrote: post #561
sokoloff wrote: post #107

Earlier quoted context omitted.

If I brake check someone on the highway, causing an accident, and it’s discovered that that’s what happened, I’d be pretty shocked* if the trailing driver was found responsible. That video looks like a combined lane change and brake check on the part of the Tesla. * and disappointed

That would still be your fault for following too closely. Basically if you hit a car from behind, it’s your fault, every time.

> Basically if you hit a car from behind, it’s your fault, every time.

Incorrect. If you brake check someone (what the Tesla did), the front car is at fault.

Re: New video of Tesla crash demonstrates the problem of semi-automated driving

#628

Post by mschuster91 »

throwaway894345 wrote: post #600
mschuster91 wrote: post #599

Earlier quoted context omitted.

> I don’t think “everyone should be allowed to use the road safely” is a statement anyone disagrees with, but the laws of physics make this very difficult to implement in any practical way. Pretty sure that even in Europe, cycling and motorcycling are far more dangerous than driving. Oh, of course we can practically implement road safety for everyone: - limit inner-city speeds outside of major influx roads to 30 km/h…

Those things improve safety, and we should certainly do them, but that’s likely not enough to bring all modes of transit into parity with regard to safety.

We can get near parity though - in the end it's probably a classic pareto distribution problem. The last 20% of traffic fatalities will be really hard, I agree, but we can reduce the utter majority of traffic accidents very very easily.

Re: New video of Tesla crash demonstrates the problem of semi-automated driving

#629

Post by jjav »

alanfranz wrote: post #317
lamontcg wrote: post #201

Earlier quoted context omitted.

> In a pileup like this it's basically never the fault of the front car, unless maybe if they are purposely causing the accident for insurance fraud or something. Quebec woman who stopped for ducks, causing fatal crash, loses appeal https://www.cbc.ca/news/canada/montreal/emma-czornobaj-loses...

That’s an highway, you cannot randomly stop there. She didn’t crash into ducks.

> That’s an highway, you cannot randomly stop there.

Right, same as what the Tesla did, suddenly stopping on a highway (being in a tunnel makes it worse, as there's no runoff room to the sides).

Re: New video of Tesla crash demonstrates the problem of semi-automated driving

#630

Post by jjav »

Alex3917 wrote: post #530
scoopertrooper wrote: post #514

Earlier quoted context omitted.

It looks like the Tesla driver will be at least as fault for the collision with the car behind it. However, it seems the subsequent collisions were caused by fleshy human drivers driving unsafely. I don't know much about traffic law, so I'm unsure how responsibility for the overall pileup will be divided. From the police report: > V-1 made an unsafe lane change (21658(a) California Vehicle Code) and was slowing to a…

Right so if there was a lane change right before the Tesla stopped then it makes sense that they might have at least partial liability for the second car. But they will probably have zero liability for all the other cars.

> Right so if there was a lane change right before the Tesla stopped then it makes sense that they might have at least partial liability for the second car. But they will probably have zero liability for all the other cars.

I know someone who had to suddenly brake due to a pedestrian jumping onto the road. So, for a very good reason in that case. Nonetheless, it was a sudden stop, so their insurance had to pay for the repairs to both of the two following cars which rear ended each other.

Post reply on HN