Page 1 of 2

Uber Autonomous Car Crash

Posted: Sun Mar 26, 2017 6:53 pm
by akirk
mmm

Image
(c) Reuters

Uber has pulled its self-driving cars from the roads after an accident which left one of the vehicles on its side.
Pictures posted online showed the car on its right side on an Arizona street, next to another badly damaged vehicle.
The car - a Volvo SUV - was in self-driving mode at the time of the crash, on Friday, Uber said. No one was hurt.
A spokeswoman for the police in Tempe, Arizona said the accident occurred when another vehicle "failed to yield" to the Uber car at a left turn.


I reckon we still have a way to go...

Alasdair

Re: Uber Autonomous Car Crash

Posted: Sun Mar 26, 2017 7:23 pm
by ChristianAB
Do you remember signing in to be a guinea pig in this giant experiment ?
You don't ? ...Oh, that's unfortunate...

Re: Uber Autonomous Car Crash

Posted: Sun Mar 26, 2017 7:50 pm
by Horse
Who has the sig line about "good drivers don't have crashes, advanced drivers don't have other people's"?

Sounds like that crash shows the level of 'skill' autonomous vehicles have reached . . .

Re: Uber Autonomous Car Crash

Posted: Sun Mar 26, 2017 10:56 pm
by waremark
At the current state of the art, do the autonomous vehicles have a better or worse accident rate than ones driven by human drivers? I don't think we know.

There was a person at the wheel of this car, responsible for taking control if necessary. He/she evidently had not perceived that it was necessary.

Re: Uber Autonomous Car Crash

Posted: Mon Mar 27, 2017 6:46 am
by akirk
I suspect a better accident rate - esp. around minor dings / knocks / etc.
However, there is clearly a mismatch between PR puffery / expectation - and reality... an assumption that autonomous cars are more capable than is true, there is still no answer provided to some of the 'moral or ethical dilemmas' which have been posed (group of nuns crossing the road v. children going to school on the pavement!) - yet the implication driven by marketing from those with a commercial interest is that the technology works...
So with this accident - did the driver not intervene because they considered the situation safe, or because mistakenly they assumed that the car was in control?
Alasdair

Re: Uber Autonomous Car Crash

Posted: Mon Mar 27, 2017 7:20 am
by ChristianAB
It seems that instead of 'autonomous' they should be called 'pre-programmed' vehicles really.
From the start, the idea that a human could intervene in an emergency seemed ludicrous to me. This is starting to show why.

Re: Uber Autonomous Car Crash

Posted: Mon Mar 27, 2017 8:42 am
by jont-
ChristianAB wrote:It seems that instead of 'autonomous' they should be called 'pre-programmed' vehicles really.
From the start, the idea that a human could intervene in an emergency seemed ludicrous to me. This is starting to show why.

Humans don't always get it right either. But generally I'm in agreement on human hand back - and Ford agree - they plan to skip level 3
https://corporate.ford.com/innovation/a ... -2021.html

There's a bit more comment here about the uber incident.
http://www.eetimes.com/document.asp?doc_id=1331523
but at the moment it seems there's not enough information to draw any conclusions.

Re: Uber Autonomous Car Crash

Posted: Mon Mar 27, 2017 9:48 am
by IcedKiwi
waremark wrote:At the current state of the art, do the autonomous vehicles have a better or worse accident rate than ones driven by human drivers? I don't think we know.

I think it's probably better but we hear about every single incident. Similarly there's 00's of internal combustion engine vehicle fires across the US every day but if a single Tesla goes up in smoke it makes the news

Re: Uber Autonomous Car Crash

Posted: Mon Mar 27, 2017 10:39 am
by ancient
Updates to correct the response (to a car failing to yield) will no doubt be made to all (Uber) cars (which is not the case when meat computers make the same mistake).
akirk wrote:I suspect a better accident rate - esp. around minor dings / knocks / etc.
However, there is clearly a mismatch between PR puffery / expectation - and reality... an assumption that autonomous cars are more capable than is true, there is still no answer provided to some of the 'moral or ethical dilemmas' which have been posed (group of nuns crossing the road v. children going to school on the pavement!) - yet the implication driven by marketing from those with a commercial interest is that the technology works...
So with this accident - did the driver not intervene because they considered the situation safe, or because mistakenly they assumed that the car was in control?
Alasdair

In what way is that moral dilemma different from when you (an AD) encounter the same situation? What would your response be?

My response to the developing possibility of pedestrians (nuns or otherwise) crossing the road without taking into account the presence of my moving car, whilst 'escape' options are reduced by (for example) a full pavement (whether of children or more nuns), would be to slow enough that I could stop if they actually step out in front of me; but us meat computers do all make different choices ;) .

Re: Uber Autonomous Car Crash

Posted: Mon Mar 27, 2017 11:26 am
by angus
waremark wrote:There was a person at the wheel of this car, responsible for taking control if necessary. He/she evidently had not perceived that it was necessary.


Typical Uber driver then :P