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

Man accused of dangerous driving for sleeping in self-driving, speeding Tesla

globalnews.ca

Post reply on HN
Print view View original

71–80 of 93 posts

Re: Man accused of dangerous driving for sleeping in self-driving, speeding Tesla

#71

Post by bb88 »

AnthonyMouse wrote: post #48
Cro_on wrote: post #11

I find it a little concerning that the car is allowed to drive in autopilot with the seats reclined. I thought there were warnings for taking your hands off the steering wheel? I suppose they can be ignored. But full recline is a bit OTT.

A car is not your mother. It has no obligation to prevent you from speeding or driving the wrong way down a one way street or otherwise being an idiot. And it shouldn't try, because sometimes the car has an inaccurate map or doesn't know that something unusual is happening or the sensor is broken etc.

> A car is not your mother.

Rhetoric like this works until it won't anymore. And that rhetoric breaks when innocent bystanders start getting killed in numbers. Line up enough grieving families in front of congress and congress will act, and then the new rhetoric will be "The car should be your mother."

Re: Man accused of dangerous driving for sleeping in self-driving, speeding Tesla

#72

Post by speedgoose »

wayeq wrote: post #40
speedgoose wrote: post #10

One solution is to put a camera inside to monitor the driver behaviour. Mazda is already selling cars with this technology.

Tesla makes you apply force to the steering wheel occasionally, but there are hacks that to try to circumvent it.

Yes it's too easy to circumvent it. The camera solution is a lot safer.

Re: Man accused of dangerous driving for sleeping in self-driving, speeding Tesla

#73

Post by maxerickson »

AnthonyMouse wrote: post #48
Cro_on wrote: post #11

I find it a little concerning that the car is allowed to drive in autopilot with the seats reclined. I thought there were warnings for taking your hands off the steering wheel? I suppose they can be ignored. But full recline is a bit OTT.

A car is not your mother. It has no obligation to prevent you from speeding or driving the wrong way down a one way street or otherwise being an idiot. And it shouldn't try, because sometimes the car has an inaccurate map or doesn't know that something unusual is happening or the sensor is broken etc.

I mean, there's interlocks between the shift mechanism and brakes, at least in part because it's safer.

Which there's two things to separate out; the lack of a feature on a vehicle doesn't particularly free the driver from responsibility for things they do, and then there's what sort of idiotic vehicles we are willing as a society to share the road with.

"A car is not your mother" doesn't really answer the second part.

Re: Man accused of dangerous driving for sleeping in self-driving, speeding Tesla

#74

Post by phaemon »

robocat wrote: post #39
anonymousiam wrote: post #26

Earlier quoted context omitted.

I have told my kids many times that I expect piloted cars to be a rarity in the future. Once reliable self-driving cars are common, the difficulty in obtaining a driver license will likely increase.

It’s entirely possible that self-driving cars could be realistic about the time we all get flying cars.

The reason we want self driving cars is the same reason we don't have flying cars.

Re: Man accused of dangerous driving for sleeping in self-driving, speeding Tesla

#75

Post by fiddlerwoaroof »

orange_tee wrote: post #49
fiddlerwoaroof wrote: post #45

Earlier quoted context omitted.

I was highly skeptical of these systems too: for normal freeway driving, Tesla’s systems are much better than I expected them to be. There’s no way I would be comfortable enough to sleep in my car, as a programmer, but I could imagine someone getting overconfident. This has always been a major risk with semi-automated systems: I believe airplanes dialed back the functionality of autopilots to prevent pilot complacenc…

OK that is a good point. So you are saying, maybe the marketing wasn't what misled the driver, but the abilities of the system. A system that fails 1% of the time will give a the user overconfidence up until the first time it fails.

Yeah, this is a good way to put it.

Re: Man accused of dangerous driving for sleeping in self-driving, speeding Tesla

#76

Post by fiddlerwoaroof »

gambiting wrote: post #53
fiddlerwoaroof wrote: post #37

Earlier quoted context omitted.

My Tesla has regularly come to a full stop from freeway speeds in response to traffic ahead. This may not be 100% reliable, but it’s not true that “the system is trained to ignore stationary objects”. It also beeps when it thinks I’m going to hit a stopped car and other things that demonstrate that this claim is false.

The question was specifically about this situation as described in the article - the car moving at 150km/h. No, your Tesla wouldn't stop in time for a fully stopped car in that situation. It does slow down for traffic, but I think many people don't realize something - it works with moving traffic, because then it can definitely recognize that you are approaching a car(or at least something that moves in the same dire…

>> These are not my words, that's exactly what Tesla said after the "trailer across a highway" accident

>> Read up on it, I'm sure the upper limit for this function is when the delta speed is I think both of these have the same explanation: if you want to release a feature like this specced at a 50kph delta, you design for a safety factor of 2-3 (100-150) so that you can be confident that it’s safe at 50kph. The claim that “it’s physically impossible” doesn’t make sense to me: humans drive safely with such deltas using only “a video feed” and sound.

Anyways, I’m fairly certain I’ve come to a stop on autopilot from at least 60mph (100kph).

Re: Man accused of dangerous driving for sleeping in self-driving, speeding Tesla

#77

Post by threatofrain »

bb88 wrote: post #71
AnthonyMouse wrote: post #48

Earlier quoted context omitted.

A car is not your mother. It has no obligation to prevent you from speeding or driving the wrong way down a one way street or otherwise being an idiot. And it shouldn't try, because sometimes the car has an inaccurate map or doesn't know that something unusual is happening or the sensor is broken etc.

> A car is not your mother. Rhetoric like this works until it won't anymore. And that rhetoric breaks when innocent bystanders start getting killed in numbers. Line up enough grieving families in front of congress and congress will act, and then the new rhetoric will be "The car should be your mother."

Then IMO it doesn't seem too invasive to have a breathalyzer test attached to all vehicles, which will certainly save a lot of lives. The data can be all local and regularly wiped to mitigate the privacy concern.

Are the numbers behind alcohol-related vehicular injuries and deaths not high enough to be a political or public priority?

Re: Man accused of dangerous driving for sleeping in self-driving, speeding Tesla

#78

Post by gambiting »

fiddlerwoaroof wrote: post #76
gambiting wrote: post #53

Earlier quoted context omitted.

The question was specifically about this situation as described in the article - the car moving at 150km/h. No, your Tesla wouldn't stop in time for a fully stopped car in that situation. It does slow down for traffic, but I think many people don't realize something - it works with moving traffic, because then it can definitely recognize that you are approaching a car(or at least something that moves in the same dire…

>> These are not my words, that's exactly what Tesla said after the "trailer across a highway" accident >> Read up on it, I'm sure the upper limit for this function is when the delta speed is I think both of these have the same explanation: if you want to release a feature like this specced at a 50kph delta, you design for a safety factor of 2-3 (100-150) so that you can be confident that it’s safe at 50kph. The clai…

>>The claim that “it’s physically impossible” doesn’t make sense to me: humans drive safely with such deltas using only “a video feed” and sound.

My logic is as follows - at 150km/h, you are covering 41m per second, and an approximate stopping distance from that speed is about 130m. Human eyes are much better at recognizing objects from a distance than computer based vision is, and Tesla is in fact relying on cameras for its forward object detection, plus a rudimentary distance-based radar. There is no chance(that's why I said "physically impossible") that whatever camera is mounted in the Tesla can reliably recognize an object(and tell that it's stopped!) at 130m. Of course the system needs to do the processing, make a decision, send a signal to the brake actuators and actually engage them. Let's be generous and add a full second to this - so to stop from that speed Tesla would need to recognize a car, identify it as a hazard, and make a critical "all brakes at maximum strength" decision from 170m away. There's no chance.

>>Anyways, I’m fairly certain I’ve come to a stop on autopilot from at least 60mph (100kph).

Ok, but there will be an upper limit to this, and I'd love to know what it is. I know that Deimler's solution only guarantees full avoidance at deltas up to 50km/h, and "reduced" impact at higher deltas - it just doesn't see far enough. Tesla's technology is fundamentally very similar, so I'd love to know what they consider as reasonable distance for full autonomous stop.

Re: Man accused of dangerous driving for sleeping in self-driving, speeding Tesla

#79

Post by manicdee »

bb88 wrote: post #71
AnthonyMouse wrote: post #48

Earlier quoted context omitted.

A car is not your mother. It has no obligation to prevent you from speeding or driving the wrong way down a one way street or otherwise being an idiot. And it shouldn't try, because sometimes the car has an inaccurate map or doesn't know that something unusual is happening or the sensor is broken etc.

> A car is not your mother. Rhetoric like this works until it won't anymore. And that rhetoric breaks when innocent bystanders start getting killed in numbers. Line up enough grieving families in front of congress and congress will act, and then the new rhetoric will be "The car should be your mother."

There are more accidental deaths from firearms than Teslas. Clearly we need better regulation of firearm handling and ownership.

Re: Man accused of dangerous driving for sleeping in self-driving, speeding Tesla

#80

Post by Isinlor »

gambiting wrote: post #44
jessaustin wrote: post #35

Earlier quoted context omitted.

Aren't there fully self-driving vehicles undergoing testing on normal city streets with stoplights? We could say that's a different mode than high-speed highway driving, but there are a number of locations where those modes blend into each other. Are the experimenters extra careful to avoid those locations? How long before the tech exists to address this issue?

Sure there are, and yes, they are different sets of problems though. Have a look at the British Tesla Driver Youtube channel, some of his videos are eye opening. Basically the car is in full autopilot mode, approaches an intersection, correctly slows down, waits for its turn, starts moving.....and in the middle of the turn goes BEEP BEEP BEEP and disengages entirely because the road markings aren't there and it wasn'…

It's an active research field. E.g. from October 2020: Calibrating Deep Neural Networks using Focal Loss: https://arxiv.org/abs/2002.09437

> Miscalibration - a mismatch between a model's confidence and its correctness - of Deep Neural Networks (DNNs) makes their predictions hard to rely on. Ideally, we want networks to be accurate, calibrated and confident. We show that, as opposed to the standard cross-entropy loss, focal loss [Lin et. al., 2017] allows us to learn models that are already very well calibrated. When combined with temperature scaling, whilst preserving accuracy, it yields state-of-the-art calibrated models. We provide a thorough analysis of the factors causing miscalibration, and use the insights we glean from this to justify the empirically excellent performance of focal loss. To facilitate the use of focal loss in practice, we also provide a principled approach to automatically select the hyperparameter involved in the loss function. We perform extensive experiments on a variety of computer vision and NLP datasets, and with a wide variety of network architectures, and show that our approach achieves state-of-the-art calibration without compromising on accuracy in almost all cases.

Calibration will be practically solved in couple of years. Then a bit longer for addressing adversarial robustness.

Post reply on HN