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

Tempe Police Release Video of Uber Accident

twitter.com

Post reply on HN
Print view View original

991–1000 of 1001 posts

Re: Tempe Police Release Video of Uber Accident

#991

Post by thisacctforreal »

agumonkey wrote: post #866
jackpirate wrote: post #78

Earlier quoted context omitted.

It's clear from the video that a human driver actually would've had more trouble since the pedestrian showed up in the field of view right before the collision, yet that's in the visible spectrum. That's not at all clear to me. I don't know too much about cameras, but it looks to me like the camera is making the scene appear much darker than it actually is. In the video, you can see many street lights projecting down…

The driver wasn't fixed on the road but he glanced two times prior the collision and had no hesitation. It seems that it was at least dark enough for him not to notice a person + a bike on a large road. I'm also very (sadly) surprised that she crossed that kind of road at night without hurrying or reacting to the sound of cars approaching.

However, the interior seemed filled with light, and they didn't glance up long enough for their eyes to adjust properly.

Re: Tempe Police Release Video of Uber Accident

#992

Post by goldenkey »

The other commenters here have done a good job of examining this incident under the lens of computer vision and AI/ML automated driving.

So I would like to take a moment to remind everyone, especially pedestrians but also drivers: To ignore the road is to play russian roulette. It doesn't matter where you grew up, even if it was a place where drivers always stopped for pedestrians and were oh so polite. Even a car at a stoplight could malfunction itself or the person driving it errs... Always keep your distance and track anything heading in your direction. It is a tragedy what happened here but it begs the question, at what point of negligence do we classify risky behavior? Russian roulette is clearly suicide but crossing a street at night and not paying attention to headlights in the distance isn't? I'd say the relative risks are vastly different but the outcome similar.

What is even worse for life loving pedestrians is the fact that cars can have their headlights turned off at night(usually newer models cant but older ones can.) And then add in EV and the car is invisible and silent... Well clearly it starts to look like someone driving under these conditions is just a potential murderer. But similar scenarios are worth thinking about. Really, anyone walking the street at night should be aware of the non obvious cues of motor vehicles.

Re: Tempe Police Release Video of Uber Accident

#993

Post by boznz »

martin_bech wrote: post #779
atonse wrote: post #25

How did LIDAR and IR not catch that? That seems like a pretty serious problem. It's clear from the video that a human driver actually would've had more trouble since the pedestrian showed up in the field of view right before the collision, yet that's in the visible spectrum. When I argue for automated driving (as a casual observer), I tell people about exactly this sort of stuff (a computer can look in 20 places at t…

Im conviced 99-100% that the Automatic Emergency Brakeing (AEB) on my Tesla would have braked for that. The promise of these systems, as you also point out, is that they can see things, humans cant. The "real" cameras on this car (not this dashcam footage), and the LIDAR, should be fine with it beeing near pitch black.

I can almost guarantee every variation on the above accident is being tested now by waymo, tesla et-al..

Re: Tempe Police Release Video of Uber Accident

#994

Post by YeGoblynQueenne »

Animats wrote: post #496
atonse wrote: post #25

How did LIDAR and IR not catch that? That seems like a pretty serious problem. It's clear from the video that a human driver actually would've had more trouble since the pedestrian showed up in the field of view right before the collision, yet that's in the visible spectrum. When I argue for automated driving (as a casual observer), I tell people about exactly this sort of stuff (a computer can look in 20 places at t…

How did LIDAR and IR (?) not catch that? That seems like a pretty serious problem. Something is badly wrong there. That should have been detected by LIDAR, radar, and vision. Yes, they need a wide dynamic range camera for night driving, but such things exist.[1][2] They're available as low-end dashcams; it's not expensive military night vision technology. Radar should pick up a bicycle at that range. The old Eaton VO…

I don't understand why everyone in this thread is so focused on the sensors alone. The sensors might detect anything they like, but they're not going to stop the car on their own. The car has logic that tells it how to react to what its sensors perceive- that's the AI part. If the car's AI can't identify a woman pushing a bike as a woman pushing a bike, or it doesn't know that it has to stop before hitting her- well, then it won't.

There's so much confusion here, about the capabilities of these systems. People think that a combination of better sensory perception + faster reaction times suffices to drive in the chaos of the real world. That's not so. Sensors and fast thinking won't get you nothing if you can't think right. You have to be able to know what the things are that your sensors detect, and how to react to them.

It's perfectly possible that the Uber' car's LIDAR detected the lady crossing the road- but the AI just didn't know what to do about her and simply did nothing.

Re: Tempe Police Release Video of Uber Accident

#995

Post by yayr »

atonse wrote: post #25

How did LIDAR and IR not catch that? That seems like a pretty serious problem. It's clear from the video that a human driver actually would've had more trouble since the pedestrian showed up in the field of view right before the collision, yet that's in the visible spectrum. When I argue for automated driving (as a casual observer), I tell people about exactly this sort of stuff (a computer can look in 20 places at t…

There is already some analysis of the footage, that claims the police video is manipulated, e.g.: https://www.youtube.com/watch?v=sKBy4_3azBg

also there are videos showing the real conditions with simple smartphone cams, e.g.: https://www.youtube.com/watch?v=1XOVxSCG8u0

Imo this is a clear attempt to shift public opinion. Also for everybody it should be clear, that a car should only drive as fast to be able to break once something comes into sight. For the whole industry it is very unfortunate, that some just don't understand the responsibility that comes with new technology and it's limitations.

Re: Tempe Police Release Video of Uber Accident

#996

Post by Arzh »

jackpirate wrote: post #78
atonse wrote: post #25

How did LIDAR and IR not catch that? That seems like a pretty serious problem. It's clear from the video that a human driver actually would've had more trouble since the pedestrian showed up in the field of view right before the collision, yet that's in the visible spectrum. When I argue for automated driving (as a casual observer), I tell people about exactly this sort of stuff (a computer can look in 20 places at t…

It's clear from the video that a human driver actually would've had more trouble since the pedestrian showed up in the field of view right before the collision, yet that's in the visible spectrum. That's not at all clear to me. I don't know too much about cameras, but it looks to me like the camera is making the scene appear much darker than it actually is. In the video, you can see many street lights projecting down…

I think you are wrong since this people die of this exact scenario almost everyday. The camera might be making it darker but that doesn't mean that every driver (Everyone's eyes and reaction times are different) would have been able to see her and get out of the way.

Re: Tempe Police Release Video of Uber Accident

#997

Post by TheCoelacanth »

eric_h wrote: post #478
jackpirate wrote: post #78

Earlier quoted context omitted.

It's clear from the video that a human driver actually would've had more trouble since the pedestrian showed up in the field of view right before the collision, yet that's in the visible spectrum. That's not at all clear to me. I don't know too much about cameras, but it looks to me like the camera is making the scene appear much darker than it actually is. In the video, you can see many street lights projecting down…

The whole reason many people on here have been advocating for self driving cars is that they can see obstructions more or less perfectly in the dark with LIDAR. I am much more interested in what that sensor said. I'm reluctant to infer exactly what a human eye would have seen in that situation. I have absolutely driven down streets in suburbia where the gap between street lights was large enough to make them quite da…

Preventing the accident might not have been possible, but even being able to decrease speed by a tiny amount would have greatly improved the pedestrian's chance of survival. Slowing from the 38mph that the car was traveling down to 30mph would decrease the chance of fatality from about 45% to below 10%.

Re: Tempe Police Release Video of Uber Accident

#998

Post by linsomniac »

JoeAltmaier wrote: post #956
linsomniac wrote: post #953

Earlier quoted context omitted.

I was under exactly the same impression. From the description a few days ago, I had imagined someone on the right hand side of the road stepping out from between two parked cars. From the video this morning I was wondering how it was possible the vehicle lidar and radar didn't pick this up. This is exactly the sort of thing I would hope these additional sensors could pick up easily. Also from the released video, it w…

Yes, particularly the 'theatre' part. Its expensive and useless. More useful would be to have a central location with humans monitoring dozens of cars, like a sort of air-traffic control situation. Better chance of keeping their attention with enough going on. They'd be there in case of difficulties, know when a car 'went off the rails' or notify authorities in case of an accident.

And with Central monitoring you could easily have people switch out frequently, and run drills like they do with lifeguards. The lifeguards at our neighborhood pool change positions every 15 or 30 minutes or go on break, with new ones cycling in. The water park does something where the supervisors will periodically introduce a special ball to the water which the lifeguards have to "save".

I worry that monitoring dozens of cars would be too much information for the "Road Traffic Control" to be able to respond in the 1.5 seconds that were available in this video, especially if they have such "low dynamic range" as what we see in this video. But maybe Lidar data would have showed the RTC operator something the car didn't see?

I'm really looking forward to the findings about what radar, lidar, and sonar sensors were saying during this time.

Re: Tempe Police Release Video of Uber Accident

#999

Post by stolsvik »

badestrand wrote: post #96
darkerside wrote: post #79

Earlier quoted context omitted.

I wonder if this was affected by the fact that the pedestrian was in another lane up until the very last second. Perhaps the car detected the pedestrian but failed to consider it an obstacle since it wasn't in its direct path. It could be unexpectedly difficult to account for pedestrian crossing speed if it caused automated cars to stop when a car in the next lane happened to "wobble" towards the automated car's lane…

There is not only continuing and stopping, there is also slowing down, as many human drivers do as well in unclear situations.

This is the exact thing I've been thinking all the time: Something's not totally right here: Let's at least slow down to better assess the situation. Everything works better in slo-mo. My human brain would then have much better time to evaluate the scene. This would definitely also hold for the sensors and processing systems for an autonomous vehicle.

Re: Tempe Police Release Video of Uber Accident

#1000

Post by stordoff »

ng-user wrote: post #922
Robin_Message wrote: post #877

Why are safety drivers not: - working in pairs, so there is social pressure, conversation, and two pairs of eyes to increase alertness and safety? - doing shifts of 30 - 45 minutes at most [1] (although they could potentially swap back and forth with a co-driver) - issued a dumb-phone for emergencies and searched for entertainment devices (it's good enough for Amazon warehouse staff) - being monitored by the driver-f…

All of those suggestions will increases cost and decrease profit, there's no law mandating any of those so why would they go above and beyond for 'safety' reasons? I understand it's the ethical decision but at the end of the day for-profit companies are only interested in one thing: maximizing profits.

Robin_Message basically says a lot of what I was thinking, but you also need to consider what maximises profits in the long term (and self driving cars are a long term project) - is it doing the minimum now to reduce costs, or is it taking all reasonable precautions to avoid the public perception building that self-driving cars are unsafe?
Post reply on HN