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

761–770 of 1001 posts

Re: Tempe Police Release Video of Uber Accident

#761

Post by nraynaud »

mdorazio wrote: post #603
osrec wrote: post #592

Earlier quoted context omitted.

The lady can be seen fairly clearly (even in this poor quality video) at 0.03 and impact occurs at 0.04. That's 1s, which means a distance of approx 17m. If the guy was watching (sort of the point of him being there really) he could have slowed the car significantly and probably even stopped it. These are test vehicles being treated like prod vehicles. They should probably not be on streets with pedestrians quite so…

You're massively underestimating human response time to visual cues and also the distance needed to stop a vehicle traveling at 35-40mph. It takes a full quarter of a second to respond to a visual stimulus on average, and more than that to also move your foot and depress a brake pedal. By that time the car was less than 50 feet from the pedestrian. At 40mph braking distance is about 80 feet in good conditions. There…

You absolutely don’t need to brake in this situation, there are 5 lanes, you release the gas to give the pedestrian more time to finish crossing and if it’s a bit short you go to the next lane over, aiming behind the pedestrian. That’s what everybody was doing in France, now that I am in Arizona the drivers are just murderous towards pedestrians, they don’t release the gas or move to aim behind them, like it’s perfectly ok to kill someone. (Don’t get me started on right on red and exiting from a drive in, I think I will get killed on a sidewalk here)

Re: Tempe Police Release Video of Uber Accident

#762

Post by nraynaud »

CodeWriter23 wrote: post #301
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 pretty clear to me (from the second half of the video) the driver was looking down at her phone and glancing up at the road periodically. IMO if she had been focusing on the road, she would have at least started braking before hitting the pedestrian. Or perhaps actually stopped before that happened.

My more generous interpretation is that they were looking at the computer screen where the car shows its interpretation of the situation, people tend to lift their phone towards their face.

Re: Tempe Police Release Video of Uber Accident

#763

Post by Myrmornis »

A human driver would have braked and swerved left into the other lane; there was obviously no oncoming traffic.

A human driver would have seen the woman much better than this camera footage makes out.

The Uber attendant was sadly not paying attention, which is going to look very bad for the self-driving car industry, even if it is very hard to ask humans to pay attention and do nothing.

Re: Tempe Police Release Video of Uber Accident

#764

Post by nraynaud »

Gatsky wrote: post #746
cameldrv wrote: post #389

Pathetic and sad performance by the vehicle and "safety" driver. The woman does not "appear out of nowhere", she was in the roadway for some time. The woman was not wearing all black, had red hair, and her shoes were reflective. Even if we are to believe that their camera is this crappy, they still have the lidar, and it appears brakes were not applied. Even 500ms of braking * 0.8g = 9mph. That might have saved her l…

What I don't understand is why the pedestrian didn't see or hear the car. She seems totally oblivious. Surely she would have seen the headlights. It is possible she had some sensory impairment, either visual or auditory.

My guess is that she felt so obviously visible that she didn’t expect to be hit. From where she started crossing she had a view of all the incoming cars for the entire length of the bridge. She was probably expecting the car to just descelerate from afar to let her cross.

Re: Tempe Police Release Video of Uber Accident

#765

Post by dotancohen »

sitkack wrote: post #417
fudged71 wrote: post #251

I do wonder if perhaps Uber made the video darker before releasing. Or used a camera that is not the main sensing camera. Because I would hope the cars are smart enough to increase ISO to get more detail at night time.

1) It does NOT show what the LIDAR sees. 2) They should be using a dark adapted / night vision camera and probably are, the video feed presented is just that, the best one that "presents". 3) They should be using thermal infrared to spot living things 4) They were severely over driving their headlights if what we are seeing is "reality" as seen by the onboard computer. I want to know how much access to the hardware U…

> They were severely over driving their headlights if what we are seeing is "reality" as seen by the onboard computer.

This is the salient point that I was looking for.

Interestingly, US and European headlights standards are very, very different and this accident is a clear example. US headlights are intended to "illuminate the roadway", whereas European headlights are intended to "illuminate obstacles on the road and in the margins". That's not just a wording difference, rather the part numbers for US-spec and European-spec headlight assemblies are different.

Re: Tempe Police Release Video of Uber Accident

#766

Post by drawkbox »

cozzyd wrote: post #574
drawkbox wrote: post #542

Earlier quoted context omitted.

Arizona and Tempe especially have lots of darker roads. The LIDAR/computer vision tuning at night doesn't seem right. Maybe it was adjusting for the changing street light brightness but yes this is one situation and the vehicle had just come from the Mill bridge that has festive lights that are strung along the bridge [1]. This is a situation though where the LIDAR should have clearly been better than it was. Maybe i…

What the heck are those trails doing in the median if it's not a legal crosswalk?

Lots of parks around and trails that do go across the road, it is an odd area.

If you zoom out on google maps you will see some of the trails. Note the sidewalk/pathway, it is no pedestrian but has paths for them so it sends mixed signals.

Re: Tempe Police Release Video of Uber Accident

#767

Post by TheGrumpyBrit »

TylerE wrote: post #444
sitkack wrote: post #417

Earlier quoted context omitted.

1) It does NOT show what the LIDAR sees. 2) They should be using a dark adapted / night vision camera and probably are, the video feed presented is just that, the best one that "presents". 3) They should be using thermal infrared to spot living things 4) They were severely over driving their headlights if what we are seeing is "reality" as seen by the onboard computer. I want to know how much access to the hardware U…

It's Arizona. I just checked. The high today was 92F. Pavement would almost certainly be warmer than that. How's a thermal camera supposed to detect 98F on a background that is proably +/- 5 degrees of that?

If a thermal camera can't detect a person on the road, then a thermal camera is the wrong camera to use. The exact technology isn't important - the important thing is, the vehicle failed to detect a human being crossing in front of it.

Re: Tempe Police Release Video of Uber Accident

#768

Post by InclinedPlane »

Someone recorded driving the same section of road at night using a cell phone camera and you can see how much better that camera reproduces the real-world visibility of the situation: https://youtu.be/1XOVxSCG8u0?t=26

That stretch of road was not particularly poorly lit, the pedestrian would have been abundantly visible to anyone paying attention, and it should have been picked up by the Uber vehicle's vision system as well, let alone their sensor suite.

I'm going to go out on a limb and guess that the pedestrian that got hit has crossed at that area many times in the past. There is actually a pedestrian crossing on the median there, but it is purely decorative and posted with signage saying not to use it. However, the area has pretty good sight lines down the road and it does seem to be well lit, so a pedestrian who cavalierly crosses through traffic would likely nevertheless have a very low chance of getting hit.

Re: Tempe Police Release Video of Uber Accident

#769

Post by hiisukun »

Since I didn't find it in a Ctrl+F, I'll chime in to say that one of the advantages of self-driving cars is the scrutiny that accidents will come under, and how the results of that investigation can improve the driving of all other self-driving cars.

While many fatal car accidents are scrutinized to an appropriate level throughout the various countries across the world, many are not. And when they are examined in detail and the vehicle had a fault, the systems in place to issue recalls and defects are quite robust and reasonably effective, although time consuming. When the driver is at fault, the local judicial system (if it exists) can attempt to rectify that behaviour through the usual means - fines, imprisonment, training, whathaveyou.

But with a self-driving car, the autonomous systems can be updated after a collision like this is understood, improved, and distributed to existing vehicles at a scale and rate far surpassing anything possible with humans driving cars individually in every city across the world.

I don't think self-driving cars are flawless in concept or mandatory for all circumstances, but having been to fatal car accidents that are eerily similar to one another on a few too many occasions, I can welcome large scale discussions on what went on in a death such as this one - because the positive outcomes can be replicated elsewhere, hopefully saving someone from the same fate.

Re: Tempe Police Release Video of Uber Accident

#770

Post by confiscate »

nraynaud wrote: post #764
Gatsky wrote: post #746

Earlier quoted context omitted.

What I don't understand is why the pedestrian didn't see or hear the car. She seems totally oblivious. Surely she would have seen the headlights. It is possible she had some sensory impairment, either visual or auditory.

My guess is that she felt so obviously visible that she didn’t expect to be hit. From where she started crossing she had a view of all the incoming cars for the entire length of the bridge. She was probably expecting the car to just descelerate from afar to let her cross.

no she didn't expect cars to decelerate from afar. She just didn't care

She didn't even look at the road at all, not even during the very moment of impact. She was looking the other way 100% of the time right up until the crash. She didn't even care enough to look at the road at all because "looking at the road" was WAY too much work for her.

Pretty sure she didn't bother to do some kind of "physics deceleration analysis" before she crossed the road LOL

Post reply on HN