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

NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction

techcrunch.com

Post reply on HN
Print view View original

111–120 of 327 posts

Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction

#111

Post by frgtpsswrdlame »

fmihaila wrote: post #105
frgtpsswrdlame wrote: post #91

Earlier quoted context omitted.

Editted to respond to edit. (perhaps my original comment was a bit brash as well) The problem here is that you're mis specifying the null hypothesis. We have had human drivers for 100 years. New automated systems start to come into the fray. We're trying to figure out if they are safer than humans. The null in this case is that they are not. That's what we assume until proven otherwise.

> We're trying to figure out if they are safer than humans. No, we're not. We're trying to figure out if TACC has any impact either way. Accordingly, the null hypothesis is that they are no more and no less safe than human drivers alone. You need data to show the direction of the change. You just picked a direction (i.e., towards worse) that was more intuitive to you.

Fair enough. I'd still say that invalidates your original comment.

>I think it is reasonable to assume that Tesla TACC is not less safe than Tesla non-TACC, therefore the conclusion is also relevant to comparisons between Tesla level 0 (entirely human-driven, no assist) vs. Tesla level 2.

Is it really reasonable to assume it's not less safe in this case? Neither of us can say either way, seems like it would be impossible to assume either way.

Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction

#112

Post by OrwellianChild »

gwern wrote: post #99
sxp wrote: post #23

The 40% number isn't very informative. The report has multiple notes about it: ODI analyzed data from crashes of Tesla Model S and Model X vehicles involving airbag deployments that occurred while operating in, or within 15 seconds of transitioning from, Autopilot mode. Some crashes involved impacts from other vehicles striking the Tesla from various directions with little to no warning to the Tesla driver. ODI analy…

Reading it, I think it is essentially a 'before October 2015' vs 'after October 2015' comparison of overall accident rates, given the caveats there. So the -40% here should be a lower bound, since it includes a time period with 0% Autopilot usage (before October 2015) and a time period with an unknown but This strikes me as suspiciously high: what fraction of airbag-worth accidents are supposed to happen on the highw…

This accident reduction is driven not by the auto-driving part of the package, but the emergency response part. The autonomous breaking and evasive steering that occurs when an accident is imminent is what is preventing accidents.

Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction

#113

Post by yohui »

Vik1ng wrote: post #58
snewman wrote: post #2

Tesla comes off extremely well in this report. For one thing, the 40% statistic cited in the headline appears to be well supported by the NHTSA report (section 5.4) and actually manages to frame the incident in a very positive light: ODI analyzed mileage and airbag deployment data supplied by Tesla for all MY 2014 through 2016 Model S and 2016 Model X vehicles equipped with the Autopilot Technology Package, either in…

Just funny that the 40% from Autosteer seem to exactly match with the general AEB safety improvement rate also mentioned in the report... IIHS research shows that AEB systems meeting the commitment would reduce rear-end crashes[emphasis added] by 40 percent. IIHS estimates that by 2025 – the earliest NHTSA believes it could realistically implement a regulatory requirement for AEB – the commitment will prevent 28,000…

For those unaware, "AEB" stands for "Autonomous Emergency Braking". See: https://en.wikipedia.org/wiki/Collision_avoidance_system

Did Tesla cars have AEB prior to Autopilot installation? If not, then this suggests the 40% reduction in crashes may simply be due to the installation of an AEB system. What effect Autopilot's other features may have would remain uncertain.

Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction

#114

Post by stale2002 »

frgtpsswrdlame wrote: post #89
zaroth wrote: post #76

Earlier quoted context omitted.

Which is exactly how you would want the analysis to be done. Because of course there were no autopilot miles in the before bucket you would be skewing the analysis if you only looked at autopilot miles in the after bucket. What we know is after installing the autopilot feature, accidents with airbag deployments reduced 40%. Not just during so-called "safe" autopilot miles, but over all miles driven. Now it's plausibl…

Except there are "autopilot" miles in the before bucket. They are comparing before and after Autosteer so Autosteer is in the after bucket only. TACC however is in both buckets, aka this is supposed to be a comparison showing the increase in safety when going from level1 to level2 systems but both buckets have tons of human driven miles mixed in.

Level 1 = cruise control. All cars in the entire world are level 1 autonomous these days.

Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction

#115

Post by tim333 »

mikeash wrote: post #24
stale2002 wrote: post #6

Oh, hey, will you look at that. The imperfect, incomplete, beta, level 2 self driving cars that were supposed to be the "dangerous" area of self driving are ALREADY better than human drivers. Can we stop the politics and deploy all the real self driving cars to the road immediately, since the government has proven that even the shitty variety is safer than humans?

Level 3 is the real danger area. That's the level where the driver is able to do other things, but must be prepared to retake control at any time. Those two ideas are basically contradictory: if you're not already involved with driving, it takes too long to get back into it to do anything useful. Tesla's system is level 2, meaning that the driver must (as in, is supposed to) remain engaged and aware of the driving ta…

The only level 3 cars that I think about to hit the road are Volvos and they are supposed to pull over and park if you're asleep when things are tricky. https://www.youtube.com/watch?v=asKvI8ybJ5U&feature=youtu.be...

Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction

#116

Post by kolbe »

pavlov wrote: post #39
kolbe wrote: post #16

Earlier quoted context omitted.

on the other hand, i am easily 3 times safer than the average driver, so self driving vehicles would endanger me.

Dunning-Kruger effect on display?

maybe by you and your command of being able to judge commenters on hacker news.

Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction

#117

Post by krschultz »

dkonofalski wrote: post #8

I don't really know why this is surprising. Computers are already better than humans at most tasks that involve a limited set of behaviors and they have infinitely better response time than humans (and continue to get better). How could anyone think that a report like this was going to end up any differently?

The history of self driving cars shows that it is not an easy task. We've been working on this for over 15 years. I was on a self driving car team during 2008-2010. We won multiple intercollegiate competitions. At that time I was certain it would be 20+ years before we had something that could drive on the roads. The progress has been spectacular.

Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction

#118

Post by Ensorceled »

up_and_up wrote: post #37
Tepix wrote: post #11

Earlier quoted context omitted.

The rates dropped by 40%. The drivers remained the same. Why do you think a different driver demographic will not show a similar improvement?

Do you think peoples ability to behave/drive responsibly vs recklessly could be at all linked to their demographic profile? I am talking statistically, not looking for anecdotes...

Car insurance, and their actuarial models, are based entirely on this being true.

Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction

#119

Post by DonHopkins »

Dan_Nguyen wrote: post #72
Pxtl wrote: post #53

Earlier quoted context omitted.

There is zero chance I'd stay awake behind the wheel of a level 2 or 3 vehicle.

I'm in the same boat as you here. Level 2 and 3 are unsafe for me as an individual. When I drive, actively focusing on the road and driving keeps me awake. When I'm a passenger, the lack of mental concentration plus the gentle rocking motion of a moving car knocks me out within ten minutes. The current Tesla autopilot is obviously working well for many people, but for myself I have to wait until level 4. Anything mor…

What kind of "CARTCHA" challenges and games might be fun to play, to prove to your car that you were paying attention to the road and traffic, yet not distract you from being ready to take over driving?

Computer vision and speech recognition could be useful for proving the driver's awake, as well as driving the car itself.

https://en.wikipedia.org/wiki/Car_numberplate_game

Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction

#120

Post by Pyxl101 »

mikeash wrote: post #80
t112512512 wrote: post #73

Earlier quoted context omitted.

Does Level 3 require instant attention? I thought it allowed the car to require some parameters (such as road type or weather). If my car can handle the hours-long uninterrupted highway sections of a road trip, yet still require manual driving on city streets, that is still a huge value proposition for time & safety.

It doesn't require instant attention, no. But it's hard envision a scenario where the car needs attention at some point in the future, not particularly soon, but can't drive itself. Maybe I'm drawing the line at the wrong point, though. If a car can drive itself but may occasionally need to pull over and ask the driver to tell it which way to go, at the driver's leisure, is that level 3 or level 4? I thought 4, but I…

Let's say that the car is entering a fog or other weather condition, which it realizes is beginning to obstruct its visibility or control (e.g. snow, heavy rain, wind). After visibility or control impairment has crossed a warning threshold, but before putting safety at risk, the car can alert the driver and begin the process of handing over manual control.
Post reply on HN