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

Police Say Uber Is Likely Not at Fault for Self-Driving Car Fatality in Arizona

fortune.com

Post reply on HN
Print view View original

321–330 of 459 posts

Re: Police Say Uber Is Likely Not at Fault for Self-Driving Car Fatality in Arizona

#321

Post by carlmr »

TeMPOraL wrote: post #308
danmaz74 wrote: post #300

Earlier quoted context omitted.

> This is definitely something that is likely to be difficult to teach to autonomous systems. Why? What you listed seem like conditions that should be easy enough to recognize even with simple algorithms.

Yup. An algorithm should be in much better position to recognize its confidence about what it sees and can happen, and alter the driving speed to reduce the amount of uncertain spots that fall within its stopping range. That is, unless they use neural network magic black boxes to implement the self-driving part.

>That is, unless they use neural network magic black boxes to implement the self-driving part.

Is this actually done? I thought neural networks were more or less used for the vision system to detect parts of the scene (pedestrians, stop-signs, etc). This is the only area where to me it makes sense to take the blackbox tadeoff right now.

Re: Police Say Uber Is Likely Not at Fault for Self-Driving Car Fatality in Arizona

#322

Post by carlmr »

gaelian wrote: post #297
CydeWeys wrote: post #168

Earlier quoted context omitted.

There is no brake involved with cruise control. What you are describing is bad drivers who do not use cruise control. It's actually extremely pleasant and predictable to be following a driver who is using cruise control. Following a well-regulated self-driving car would be similar.

My car absolutely uses the break when required to maintain the set cruise control speed, I've watched it happen. The car I owned before my current car didn't use the break to keep to the set cruise control speed though, so I get why you may think there is no break involved. It largely seems to depend on make and cost of the car you're driving.

I'd have said age and which assistance systems there are. In many older cars there's just no path for the CC to control the brakes.

Re: Police Say Uber Is Likely Not at Fault for Self-Driving Car Fatality in Arizona

#323

Post by sandos »

gaelian wrote: post #297
CydeWeys wrote: post #168

Earlier quoted context omitted.

There is no brake involved with cruise control. What you are describing is bad drivers who do not use cruise control. It's actually extremely pleasant and predictable to be following a driver who is using cruise control. Following a well-regulated self-driving car would be similar.

My car absolutely uses the break when required to maintain the set cruise control speed, I've watched it happen. The car I owned before my current car didn't use the break to keep to the set cruise control speed though, so I get why you may think there is no break involved. It largely seems to depend on make and cost of the car you're driving.

Our Passat uses brakes for cruise control, our older Kia Ceed did not. I actually hate when it uses the brake. The non-brake method is much better imo. Feels like such a waste to heat the brakes on a steep incline instead of just rolling with it.

Re: Police Say Uber Is Likely Not at Fault for Self-Driving Car Fatality in Arizona

#324

Post by tzs »

prawn wrote: post #258
benwilber0 wrote: post #33

Earlier quoted context omitted.

That is speeding and also the car was inattentive to potential hazards outside of its vision. Normal human drivers slow down when they see a wild lady on the median looking like she might dart into the road. Normal people slow down when there are kids playing with a ball on the sidewalk. Do these cars do that? Do they even drive the speed limit? edit: Also accidentally killing someone with your car because of minor s…

Sign in the area was actually 45 mph, so the car was 7 mph under the limit, it seems.

Yup. Here's a picture of the sign [1].

I found this based on the photo of the accident location in this article [2]. I went to Tempe, AZ on Google Maps, entered the name of the building in the background of the photo ("First Solar"), and from there it was pretty easy to find the photo location. Then it was just a matter of going backward until the speed limit sign for that road in that direction was found.

[1] https://www.google.com/maps/@33.4350531,-111.941492,3a,75y,3...

[2] https://www.reuters.com/article/us-autos-selfdriving-uber/se...

Re: Police Say Uber Is Likely Not at Fault for Self-Driving Car Fatality in Arizona

#325

Post by dingaling »

ShorsHammer wrote: post #314
r_klancer wrote: post #225

My takeaway from the comments in this thread: self-driving Ubers don't drive defensively, almost surely can't "read" pedestrian behavior, and even with lidar may never have seen Elaine Herzberg. Oh, and they speed. This sounds like straight out of the "hell" scenario for driverless cars: "we cannot accept responsibility for what happens to pedestrians and cyclists foolish enough to stray out of their designated safe…

I don't see why autonomous vehicles are so big. Virtually all travel is one person, ideally it would be a tiny 4 wheel electric with the sole passenger highly reclined. The hangover from internal combustion is real.

Even the original DARPA Grand Challenge in 2005 was won by a system installed on VW Taureg, an enormous two-tonne SUV.

Re: Police Say Uber Is Likely Not at Fault for Self-Driving Car Fatality in Arizona

#326

Post by gaelian »

carlmr wrote: post #322
gaelian wrote: post #297

Earlier quoted context omitted.

My car absolutely uses the break when required to maintain the set cruise control speed, I've watched it happen. The car I owned before my current car didn't use the break to keep to the set cruise control speed though, so I get why you may think there is no break involved. It largely seems to depend on make and cost of the car you're driving.

I'd have said age and which assistance systems there are. In many older cars there's just no path for the CC to control the brakes.

Yeah true. Seems like the more feature complete assistance systems are making their way down to even some of the more low-end brands these days. Level 1 autonomy is getting pretty old hat now I guess.

Re: Police Say Uber Is Likely Not at Fault for Self-Driving Car Fatality in Arizona

#327

Post by Casseres »

CydeWeys wrote: post #168
freehunter wrote: post #155

Earlier quoted context omitted.

It may not be programmed to speed per se, but if you’ve ever had cruise control turned on you’ll know it doesn’t keep you exactly at 55mpg. It might be 54 with the throttle applied to speed up, or it might be 57 which doesn’t trigger the brakes yet but is coasting to a slower speed. Imagine the car in front of you cycling rapidly between the gas and brake pedal to keep the speed exactly at 55mph. One, that’s complete…

There is no brake involved with cruise control. What you are describing is bad drivers who do not use cruise control. It's actually extremely pleasant and predictable to be following a driver who is using cruise control. Following a well-regulated self-driving car would be similar.

My car has a CVT (Continuously Variable Transmission). With cruise control enabled, going downhill, it will actually change the effective gear ratio and "downshift" automatically.

(Once or twice when going downhill without cruise control, just using the brake pedal, it "downshifted" automatically while I had the brake pedal pressed.

CVTs are computer controled and better than manual/standard for gas mileage, but with the downaide that you can't tow heavy loads.)

Re: Police Say Uber Is Likely Not at Fault for Self-Driving Car Fatality in Arizona

#328

Post by franciscop »

URSpider94 wrote: post #315
in_cahoots wrote: post #158

Earlier quoted context omitted.

I don’t think that’s true at all (though I only have my own experience to go off of). I tend to use the speed limit as a point of reference, and then drive a little above it. Prior experience has told me that going the speed limit plus some percentage ‘feels’ safe. If there were no speed limit I’d actually have to pay more attention to the road and surrounding environment to determine what a safe speed is. Right now…

This is well-studied. It turns out that the 85th percentile speed (the speed where 85% of drivers are below that speed, 15% above) is nearly insensitive to speed limit. Road design is the biggest single determinant of travel speed. http://www.lsp.org/pdf/troopc85thSpeed.pdf

Amazing insights:

> Won’t raising the speed limit cause people to drive faster and cause more crashes?

> The Federal Highway Administration studied nearly 200 roads in 22 states where speed limits were raised, lowered or left unchanged. Prior to the speed limit change, 55 percent of drivers exceeded the posted speed limits. After speed limits were raised or lowered as much as 20 mph, there was a slight change in speed, but generally less than 1 mph. There were no significant changes in crashes, although crashes tended to decrease where speed limits were increased to realistic levels. Also, there was little effect on speeds or crashes on intersecting or nearby roadways.

Re: Police Say Uber Is Likely Not at Fault for Self-Driving Car Fatality in Arizona

#329

Post by ckinnan »

ShorsHammer wrote: post #314
r_klancer wrote: post #225

My takeaway from the comments in this thread: self-driving Ubers don't drive defensively, almost surely can't "read" pedestrian behavior, and even with lidar may never have seen Elaine Herzberg. Oh, and they speed. This sounds like straight out of the "hell" scenario for driverless cars: "we cannot accept responsibility for what happens to pedestrians and cyclists foolish enough to stray out of their designated safe…

I don't see why autonomous vehicles are so big. Virtually all travel is one person, ideally it would be a tiny 4 wheel electric with the sole passenger highly reclined. The hangover from internal combustion is real.

As autonomy is implemented and refined, and safety improves dramatically, new regulatory and economic models will emerge. This will spark design innovation. No doubt both single rider and shuttle-bus options will emerge, with much lighter builds as crashes decrease.

Re: Police Say Uber Is Likely Not at Fault for Self-Driving Car Fatality in Arizona

#330

Post by w0utert »

hnaccy wrote: post #245
isaachier wrote: post #234

Earlier quoted context omitted.

I think you might be confusing "defensively" with perfectly. There are situations even defensive driving cannot prevent. For example, any car going 35 mph will need a few feet to brake if something jumps directly in front of it. Defensive driving just accounts for one's surroundings and likely outcomes. Seeing as the cars have been on the road in Arizona for a while, and this is the first we are hearing about a pedes…

>Seeing as the cars have been on the road in Arizona for a while, and this is the first we are hearing about a pedestrian collision, it seems the algorithm is pretty safe. One fatality in 4 million miles is safe? Humans are 20 times better than that.

>> One fatality in 4 million miles is safe? Humans are 20 times better than that.

And this is considering the fatality rate for all human drivers, including those that are drunk, playing on their smartphone, sleepy, speeding, street racing, etc. I would not be surprised if those cases account for a very sizable majority of fatal car accidents. Now imagine you would reduce those numbers by relatively simple measures that don't involve AI. Let the car enforce speed limits, mandate smartphone companies to have a 'car mode' that cannot be shut off, mandate attention tracking capabilities for cars, dead-zone indicators, automatic emergency braking, etc. Maybe even mandate an alcohol lock, or at the very least a black box that records driving behavior so it can be analyzed in case of a crash, with sever penalties after the fact if the black box shows signs of drunk driving, etc.

The list of things you can do to improve road safety without requiring AI cars is endless, yet billions are invested in trying to fully automate driving. This just shows that self-driving cars are not about safety, but purely about business opportunities.

Post reply on HN