1 2 3 ... 9
ebonyandivory
ebonyandivory UberDork
3/20/18 7:45 a.m.

Hope this isn’t already on here but I thought it might be interesting to discuss.

My initial impression is I’m not ready to blame the car or the software until I know how suddenly this person appeared in front of the suv involved.

There’s already a hold put on further self-driving cars in many cities as a result.

Id hate to see an overreaction in the media (hahahaha) before we know if it was even possible for any human to have avoided it.

https://www.ndtv.com/world-news/self-driving-uber-car-kills-us-woman-in-first-autonomous-vehicle-death-1825963

camaroz1985
camaroz1985 HalfDork
3/20/18 8:29 a.m.

I agree.  I guess the good thing about AVs is that there are terabytes of data and video/lidar/radar that documented exactly what happened.  If it is something that a human could have avoided, it may slow development down (though I don't think it should), but if a human driver would have had the same outcome, then sadly this is just something that would happen regardless of who or what is in control.

I'm all for autonomous driving.  I hate the monotony of driving on highways for long periods of time.  I will keep my fun cars for weekends and racing, and gladly let the robots chauffeur me around on my daily commute and long trips.

ebonyandivory
ebonyandivory UberDork
3/20/18 8:36 a.m.

In reply to camaroz1985 :

Initial report was that the “driver”s first warning was the body impacting the car it was so sudden.

yupididit
yupididit SuperDork
3/20/18 8:37 a.m.

First one so far? How many human driver death have we had for every av out there? 

Also, what was the driver doing? They didn't see the pedestrian and override it? 

Dirtydog
Dirtydog GRM+ Memberand HalfDork
3/20/18 8:42 a.m.

Condolences to the family.  Just curious, from an insurance stand point.  How is liability determined?  Driver error?  I know witness accounts and all that.  I'm not ready for machines running my life quite yet.

STM317
STM317 SuperDork
3/20/18 8:48 a.m.

In reply to Dirtydog :

That's where it gets interesting. The autonomous cars that Uber uses (and the one involved here) are Volvos. In 2015, Volvo's CEO stated that they would assume liability when their vehicles were in autonomous mode. This vehicle was in autonomous mode, but will it be considered Volvo's autonomous tech, or Uber's?

Also, apparently under recent Arizona state law, the company operating the vehicle would be held responsible.

ebonyandivory
ebonyandivory UberDork
3/20/18 8:52 a.m.
yupididit said:

First one so far? How many human driver death have we had for every av out there? 

Also, what was the driver doing? They didn't see the pedestrian and override it? 

I mentioned above that the driver claims there wasn’t time to react. It was so sudden that the impact was the first warning. I don’t see a need to disbelieve that at this point.

1988RedT2
1988RedT2 UltimaDork
3/20/18 9:04 a.m.

Correction.  "First of many."  What I find most amazing about the discussion of "self-driving cars" is that so many people think it's just a peachy idea.  The misguided worship of technology is going to lead us down a dark road indeed.

 


 

ebonyandivory
ebonyandivory UberDork
3/20/18 9:11 a.m.

In reply to 1988RedT2 :

I think it’s awful myself. But that’s because I think driving a car isn’t a chore and shouldn’t be seen as such.

kb58
kb58 SuperDork
3/20/18 9:18 a.m.

Well zooming out and ignoring that the car was self-driving for a second, the person killed was at least half responsible if, as reported, she stepped into traffic mid-block to cross the street. Who does that and assumes the cars will stop? Yes I know it's premature to draw conclusions but I suspect the outcome would have been the same regardless what the car was. Sadly, said another way, if she'd been killed by a "normal" car, this wouldn't be news.

Titan4
Titan4 New Reader
3/20/18 9:20 a.m.

In reply to ebonyandivory :

I understand what you're saying but what about older people that just aren't up to driving any more?  The self-driving car seems like it would allow them to keep more of their independence without being as much of a threat to the rest of the world.  

rslifkin
rslifkin SuperDork
3/20/18 9:24 a.m.
Titan4 said:

In reply to ebonyandivory :

I understand what you're saying but what about older people that just aren't up to driving any more?  The self-driving car seems like it would allow them to keep more of their independence without being as much of a threat to the rest of the world.  

That's a good use for autonomous cars.  The big thing that scares me about the tech is this: if autonomous cars are determined to be much safer than human-driven ones, one of 2 things will happen.  Insurance for human-driven cars will become unaffordably expensive or driving manually on public roads will just be made illegal.  And that would suck.  

ebonyandivory
ebonyandivory UberDork
3/20/18 9:31 a.m.
Titan4 said:

In reply to ebonyandivory :

I understand what you're saying but what about older people that just aren't up to driving any more?  The self-driving car seems like it would allow them to keep more of their independence without being as much of a threat to the rest of the world.  

We have the Council On Aging for that. Van picks ‘em up at home, brings them shopping or to doctors appointments and brings ‘em home again.

My mom just had my wife fix her “broken” tv... (the volume was turned to zero). I’m pretty sure a self-driving car is out for the elderly unless it’s somehow preprogrammed for them. No thanks

 

STM317
STM317 SuperDork
3/20/18 9:48 a.m.
Titan4 said:

In reply to ebonyandivory :

I understand what you're saying but what about older people that just aren't up to driving any more?  The self-driving car seems like it would allow them to keep more of their independence without being as much of a threat to the rest of the world.  

For the last 150 years, we've had the ability to request a vehicle to come to your door, pick you up, and take you wherever you wish in exchange for money. How do autonomous vehicles solve that problem any better than a taxi?

Apexcarver
Apexcarver PowerDork
3/20/18 9:53 a.m.

In reply to STM317 :

Not paying the driver.

 

 

If I was a taxi or truck driver, well, Id be searching for a new career...

 

 

kb58
kb58 SuperDork
3/20/18 10:10 a.m.
STM317 said:

For the last 150 years, we've had the ability to request a vehicle to come to your door, pick you up, and take you wherever you wish in exchange for money. How do autonomous vehicles solve that problem any better than a taxi?

Not paying.

Not dealing with a different person each time, a person you don't know.

Predictability, not getting in an unknown car with its various smells, stains, and dirt.

Convenience, "take me to the doctor", rather than waiting around who knows how long.

Provides a sense of control over one's life, at a time when such a thing is very important.

Dirtydog
Dirtydog GRM+ Memberand HalfDork
3/20/18 10:16 a.m.

Can auto driving lead to the ultimate distracted driver?   

1988RedT2
1988RedT2 UltimaDork
3/20/18 10:45 a.m.
kb58 said:

Well zooming out and ignoring that the car was self-driving for a second, the person killed was at least half responsible if, as reported, she stepped into traffic mid-block to cross the street. Who does that and assumes the cars will stop? Yes I know it's premature to draw conclusions but I suspect the outcome would have been the same regardless what the car was. Sadly, said another way, if she'd been killed by a "normal" car, this wouldn't be news.

Aaaaand.... The rationalization begins.  So, how many accidental deaths by self-driving car are "acceptable?"  100?  100,000?  It was for the greater good.

GameboyRMH
GameboyRMH GRM+ Memberand MegaDork
3/20/18 10:49 a.m.

Not the first death in general, but the first pedestrian death. Remember the guy who drove under a semi trailer while watching a Harry Potter movie in his Tesla S on autopilot.

The numbers will work out well for autonomous driving. That's 2 deaths due to autonomous driving in total so far, out of approximately 3,287 traffic deaths per day worldwide. If the world switches to autonomous cars and has less deaths per day than that, it's still an improvement.

dculberson
dculberson UltimaDork
3/20/18 10:49 a.m.
1988RedT2 said:
kb58 said:

Well zooming out and ignoring that the car was self-driving for a second, the person killed was at least half responsible if, as reported, she stepped into traffic mid-block to cross the street. Who does that and assumes the cars will stop? Yes I know it's premature to draw conclusions but I suspect the outcome would have been the same regardless what the car was. Sadly, said another way, if she'd been killed by a "normal" car, this wouldn't be news.

Aaaaand.... The rationalization begins.  So, how many accidental deaths by self-driving car are "acceptable?"  100?  100,000?  It was for the greater good.

How many accidental deaths are acceptable in cars driven by people? We're at about 37,000 a year right now. Is that the greater good you have in mind?

codrus
codrus GRM+ Memberand UltraDork
3/20/18 10:52 a.m.
The0retical
The0retical UltraDork
3/20/18 10:55 a.m.
1988RedT2 said:
kb58 said:

Well zooming out and ignoring that the car was self-driving for a second, the person killed was at least half responsible if, as reported, she stepped into traffic mid-block to cross the street. Who does that and assumes the cars will stop? Yes I know it's premature to draw conclusions but I suspect the outcome would have been the same regardless what the car was. Sadly, said another way, if she'd been killed by a "normal" car, this wouldn't be news.

Aaaaand.... The rationalization begins.  So, how many accidental deaths by self-driving car are "acceptable?"  100?  100,000?  It was for the greater good.

A quantifiable reduction in automotive related deaths which can be directly correlated to the rise of autonomous vehicles would be my criteria. Which is unlikely to happen unless they're given more time to test.

Honestly while navigating my 3200lbs liquid dinosaur propelled missile down the road I feel the need to pay attention. Those that don't buy a 4200lbs missile because it skews the laws of physics in their favor when they're not paying attention. I'd prefer those in the latter category be a bit more predictable while they eat a bowl of soup, do their makeup, shave, or take a business call when I have as much as stake as they do. Autonomous vehicles accomplish that.

We'll see what the datalogs say in a few days about this accident. Physics applies equally to manned and autonomous cars weighing two tons and driving 45mph. So it's possible there was no reaction time.

adam525i
adam525i GRM+ Memberand Reader
3/20/18 11:11 a.m.

In reply to codrus :

Having read that article it begs the question why the UBER/Volvo car couldn't see the person even if the driver couldn't. You'd think IR cameras or at least some sort of night vision would be a must and be an obvious improvement over human eyes, it still wouldn't prevent someone from darting out from behind a wall etc but it sounds like it could have made a difference in this situation.

Adam

RevRico
RevRico GRM+ Memberand UltraDork
3/20/18 11:12 a.m.

Autonomous cars still won't fix idiot pedestrians.

 

Duke
Duke MegaDork
3/20/18 11:20 a.m.
1988RedT2 said:

Correction.  "First of many."  What I find most amazing about the discussion of "self-driving cars" is that so many people think it's just a peachy idea.  The misguided worship of technology is going to lead us down a dark road indeed.

And from what insanity I see on my daily commute on local connectors - a 22 mile round trip where speeds rarely reach 60 mph - I think self-driving cars are a peachy idea.

I'll take my chances with a high-throughput, GPS- and radar-equipped computer driving with 100% of its attention rather than a pissed off / late / stupid / distracted / phone-addicted human moron giving the task 25% at best.

1 2 3 ... 9

You'll need to log in to post.

Our Preferred Partners
pGNJGZEzqZoIqDbdw77gnPNBPxRvOYNTdLFa8gOWjqGeiVbe2drpz0y4gyfTEGZ9