Does anybody know what the "Population" field indicates at the top of the report?
NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction
231–240 of 327 posts
Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction
#232Earlier quoted context omitted.
Confounding factors. For example, crashes can be more likely after sunset or during inclement weather so seasonality can play a role. Or maybe the drivers who had Tesla longer got used to it.
Could this be accounted for if the sample period was large enough, ie over 12 months to cover seasonality and a few months for time of day.
this particular one, maybe, but there might be others. like the user base pre-ota and post ota (number-wise) could be different.
I think there is a good SaaS business with A/B testing robot/car software OTA. I would imagine some metrics would be: 1. car speed 2. battery consumption per mile. 3. user comfort metrics (if it's possible to measure, maybe like hearthrate, body temperature etc - i can imagine if robots were driving bad, it'd be visible in heart rate or something like that).4
The list probably goes on. It would be interesting to see how much battery consumption changed after OTA, for example.
Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction
#233Waiting for the headline: "Human Fails to Prevent Accident, Outraged Public Calls for Banning of all Human Drivers" The obsession with perfection in self-driving cars is misplaced, they just need to be demonstrably better than humans. This is obviously the future.
At the very least in a legal standpoint that's not true, because with a human driver it's far easier to define fault (which in turn allows for the issue to be resolved) than if two autopilots happen to collide. Who pays for what then?
From a safety point of view, I'm more inclined to agree however I'm not entirely convinced that we've passed the novelty stage of it yet where drivers will remain vigilant because it's a new and exciting feature and that mentally primes them to keep focused. It wouldn't suprise me that as it becomes more mainstream, drivers begin to be less vigilant on staying focused as society as a whole begins to trust it more. That's when everything could go bad.
Nevertheless, you're absolutely right that this is the future no matter what so we need to design it as best we can to resolve all of the issues it may give rise to.
Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction
#23440% figure is meaningless unless the absolute numbers are reported. How do we know if this difference is statistically significant ?
was 1.3 crashes / million miles. Is now 0.8 crashes / million miles.
Tesla's have driven >= 3 billion miles
Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction
#235Earlier quoted context omitted.
Confounding factors. For example, crashes can be more likely after sunset or during inclement weather so seasonality can play a role. Or maybe the drivers who had Tesla longer got used to it.
Could this be accounted for if the sample period was large enough, ie over 12 months to cover seasonality and a few months for time of day.
I suspect OTA update was applied simultaneously rather than staggered over the course of the year, so probably not. You can control for seasonality using some other data as a proxy, but the more adjustments you make, the more potential there is for introducing errors. Randomized test/control split is almost always better than before/after comparison.
Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction
#236Earlier quoted context omitted.
Ah it now makes sense, I did ask him to clarify but I just parsed in my brain when reading your reply he meant fewer wealthy (better educated drivers) as opposed to less wealthy meaning not as wealthy. Thanks for your help.
I read it as less wealthy meaning not as wealthy as well. Funny how ambiguous some sentences can be.
Well it's really that he used the wrong word. "Less wealthy drivers" can't correctly refer to the number of drivers.
Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction
#237After looking at the report it looks like Tesla ran into the same issue we did in the 2007 DARPA Urban Challenge. The trailer was higher than the front facing sensors. We and most other teams had all assumed 'Ground Based Obstacles' meant that any obstacles on the test track would make contact with the ground in the lane of travel. DARPA decided to put a railroad bar across the street and expected cars to back up and…
I don't understand why the paradigm isn't about full volumetric prediction .. is it because people were too tired of solving all other issues (surely possible at the time of DARPA challenge, but Tesla had time and money since); or hardware limits (like scanning a larger area would impede signal quality or make processing too heavy ..)
Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction
#238Earlier quoted context omitted.
The left lane is only for passing, and if you are passing >80% of the cars on the road, you are a reckless driver.
> if you are passing >80% of the cars on the road, you are a reckless driver. Or you are driving at the speed limit on a 2x2 highway used as a trucking corridor. (For example: most of I-5.)
The speed disparity can be a huge issue. Drive over the Donner Pass sometime or any major Western highway with lots of trucks and mountainous spots (i.e. most Western highways at some points) and the different speed combinations can be pretty nerve wracking to navigate.
Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction
#239Earlier quoted context omitted.
I really would like to know the absolute numbers for this reduction. It's really questionable if this would hold up with millions of Teslas on the road in the hands of less wealthy(probably better educated) drivers.
Wealthy motorists drive more recklessly: http://www.pnas.org/content/109/11/4086.short
Jerks drive Mercs
Re: NHTSA’s full investigation into Tesla’s Autopilot shows 40% crash rate reduction
#240Earlier quoted context omitted.
Would require you to link your social media accounts to your car so easily avoided.
Entice drivers to link to their social media accounts by offering features like automatically tweeting a photo of anyone you honk your horn at, and updating an online leader board of top speeds driven along every section of road.
>top speeds driven along every section of road
If your goal is to make things safer I don't think this will help.