This is just the raw camera input. The question here is, what did the software see? What did the sensors register? Is their system using an ultrasonic sensor? Did this detect anything? Commercial ultrasonic sensors can detect something at least 15' away. So the software could have at least slowed down to minimize the collision impact. What did the LIDAR see? Did their software detect an object? This accident happened…
I believe this isn't even "input", it's just a dash cam. The data available to the car hasn't been released yet.
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.
This looks to me like it came from a $30 dash cam, not the car's main visual spectrum camera sensor.
interior view has better night vision than exterior...
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…
The actual spot it happened has a bike lane, and crosswalk sidewalk by it [1].
This is such a blatant attempt to manipulate public opinion before the actual data appears. Even if Uber's software is ridiculously using visible light for its decision-making, it should definitely be taking into account what it doesn't know. A system like this should not be guessing. There is a giant swath of black in the video; the camera doesn't know what's there. So, the car should be moving so it has a short eno…
The police department released the video, not Uber.
Mm, fair enough, that's my inner conspiracy theorist happening :b (Every part of the US government is obviously in cahoots with Uber as part of a complex scheme to take over the world, right?). Okay, probably not intentional on their part, but I think it's really disappointing that this is the video they had to share. The video itself is manipulative, and surely they would realize that. It provides some context, but the prevailing focus is "look how dark much darker those pixels are than those other pixels!", which just isn't useful.
I wouldn't be one bit surprised. The antics of this company are well known. This incident has my wondering: of all the companies working on this, why is Uber trusted to be testing this technology?
trusted by _who_ in order to get approval for testing?
I'd trust someone like UL.com - we trust them with a hell of a lot, why not this?
> You'd have to see someone crossing the road like that. You could say the same thing about a car with headlights on coming towards you... From the little evidence we have it seems like the pedestrian, driver/monitor and the car all should have done better.
That pedestrian obviously didn't have good situational awareness, but are we happy to let self-driving cars mow down any unaware obstacle on the road? Broken-down cars/deer/children/roadworks? Honestly, that video proves to me that this tech is nowhere near ready. That was the most obvious of obstacles and the car did not even slow down. I believe a competent human would have at the very least slowed down in that sit…
> that video proves to me that this tech is nowhere near ready.
It proves that Ubers tech is nowhere near ready. Based on all the stupid shit I see Uber engineers saying on Blind app, I have very little faith in their maturity, professionalism, and culture.
I would bet that Google is doing a better job than Uber is.
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.
I wouldn't be one bit surprised. The antics of this company are well known. This incident has my wondering: of all the companies working on this, why is Uber trusted to be testing this technology?
In 2016 they tried just turning their autonomous vehicles loose in SF without regulatory approval. They got shut down. Rather than apply for permits, they headed off to Arizona, where the then-new governor welcomed them with open arms, an "anything-goes approach", and a lot of anti-regulation blather: https://www.nytimes.com/2017/11/11/technology/arizona-tech-i...
Wow... This is REALLY bad for self driving cars. HOW did it not catch this? I'm a huge proponent of self driving cars, but I'm shocked that the car literally did NOTHING to avoid her. I know for a fact, I would have seen and potentially swerved (at least slammed on the breaks).
2+ seconds from the time you first see the person on this really crappy video is enough time for a human to react and a car to stop from 35mph. I don't blame the driver. Trying to stay alert for hours just for an incident like this is impossible and futile. I do blame Uber's shitty tech. Even this low-res camera sees the person in time and yet it never even applies the brakes from what I can see.