Police Chief: Uber Self-Driving Car 'Likely' Not At Fault In Fatal Crash (arstechnica.com)
An anonymous reader quotes a report from Ars Technica: The chief of the Tempe Police has told the San Francisco Chronicle that Uber is likely not responsible for the Sunday evening crash that killed 49-year-old pedestrian Elaine Herzberg. "I suspect preliminarily it appears that the Uber would likely not be at fault in this accident," said Chief Sylvia Moir. Herzberg was "pushing a bicycle laden with plastic shopping bags," according to the Chronicle's Carolyn Said, when she "abruptly walked from a center median into a lane of traffic." After viewing video captured by the Uber vehicle, Moir concluded that "it's very clear it would have been difficult to avoid this collision in any kind of mode (autonomous or human-driven) based on how she came from the shadows right into the roadway." Moir added that "it is dangerous to cross roadways in the evening hour when well-illuminated, managed crosswalks are available." The police said that the vehicle was traveling 38 miles per hour in a 35 mile-per-hour zone, according to the Chronicle -- though a Google Street View shot of the roadway taken last July shows a speed limit of 45 miles per hour along that stretch of road.
Why does it look like an sidewalk?
Much like a river kills the person jumping in it?
...gis sdrawkcab (usually not responding to ACs; don't bother posting as AC)
A human driver generally would have had the intuition the the walker would or could have walked into the path. How much pronostication does AI do?
Arrest that river!
On a different topic though, what if a self-driving car kills a dog? What if it kills something we eat like a cow? What if it kills a spider? Such a question seems silly until you realize that a computer has to be programmed on what is acceptable to kill. (Even if it's "programming through training the AI".)
>> Score:2, Troll
Hmmm...not much of a sense of humor out here today I guess.
If you only hear the whoosh, please look up "Simpsons out of my way I'm a motorist"
https://www.youtube.com/watch?v=8HESH8U1B4o
There will be a thorough investigation of the vehicle, the programming, all of the data and details. Even if it is decided that the victim acted imprudently, such accidents always (at least around here, unless it was the police involved) are fully investigated, and the driver is rarely exonerated from all blame, just the proximate causal fault.
Now, for you ignats who see class discrimination in the description that the victim was pushing a bicycle laden with shopping bags, a word; the police are the upper caste in these situations. Corporations will be prosecuted more often than police officers, and more often than reputable members of the community, IE, government. Or favored citizens. This is not new.
There was more than one factor leading to this tragedy, and if the end result is change in how these vehicles monitor their surroundings to have more time to analyze and react, excellent, and if the result is a recognition that even self-driving vehicles are unable to avoid such accidents, just as even skilled and careful human drivers are, well, then we've learned that self-driving does not equal infallible. That's important, and useful information.
deleting the extra space after periods so i can stay relevant, yeah.
In my community, we have great sidewalks, many crosswalks and all that needed to create a safe and walkable community. What do the pedestrians still do, you ask?
Walk out into traffic if it's more convenient. If a car hits them after taking reasonable measures to stop, they ought to be liable for all of the damage caused including to the vehicle and the driver's therapy if required.
My wife knew someone who killed a pedestrian who just walked out into traffic like this without thinking. Totally unavoidable. The "victim" was the driver, not the pedestrian because the driver was obeying the law and some stranger decided "fuck the traffic laws" and made her party to an accidental vehicular homicide.
Traveling at 38 mph in a 35 mph zone on Sunday night, the Uber self-driving car made no attempt to brake, according to the Police Department’s preliminary investigation.
Not only was this car speeding, but it did not recognise a road side hazard and drive by cautiously.
Here in Australia we now have "Incident" laws which requires the driver to slow down to 40km/h (25miles/h). Clearly this cyclist was an incident and the uber car would definitely be at fault.
And to just openly say that if there are hazards and an autonomous vehicle doesn't recognise these then it is not the auto makers fault.
Yeah right.
The police chief needs to get some facts straight about the technology of autonomous vehicles work. LiDAR comes from LASERs. From the VEHICLE.
Unclear which "shadows" Chief Moir is talking about. Streetlights are but a one illumination source at play here.
Not just "FAKE NEWS" but OLD news too.
One advantage autonomous cars have is that even if they are found not legally at fault they don't go around feeling guilty about the incident like a person would.
Those are valid ethical questions that have to be answered by the AI and the programmers. If a car is forced to make a choice between killing a squirrel or killing a child, will it treat them the same? How will it choose if it has to decide between killing 4 people or 1?
Nah, it was just taken literally because there's a lot of SocJus propaganda out there that would say just that in all seriousness.
You forgot to put a "lol" or "=-)" at the end of your post
I modded you up although I suspect this will soon be negative.
I don't think auto-ownership is the reason though I agree with the analysis of caste.
I bet the police chief got a phone call from the mayor, something along the lines of "We get a lot of tax revenue and good will from this". You know, like the mayor in "Jaws" not wanting to close the beach.
Hmmm..not much humor here today I guess.
FTFY.
Humans can adjust to changing situations, they can also ready body language. Most people slow down when they see someone on the side of tge road looking like they are going to step out. An AI cant read that sort of thing. They can only react tl basic things presented to them.
Doesnt the car have sensors that could have detected the person and her bike with bags? Dont get me wrong it appears that the pedestrian was in the wrong but something should have been detected and the car should have done something to try and avoid the accident. Maybe these cars are not smart enough yet. Though once fully certified I would expect them to be better at driving then people. Something to work towards I guess.
Interesting series of tweets: https://twitter.com/EricPaulDe...
The median looks like it has fancy, inviting paths, but it also warns you not to use them. And the actual crossing is kind of daunting...
It is a rather bad design, but it does look dangerous in any case, so if I wanted to cross that way I would exercise extreme caution...
Violence is the last refuge of the incompetent. Polar Scope Align for iOS
A big reason why I'm still on this board is that no self-respecting SlashDot user would stoop to using emojis or first-generation Internet abbreviation in their posts. We generally grew out of that twenty years ago.
If it kills something we eat, like a cow, I think a cookout would be good.
Eat shit and die, faggot.
>> even if they are found not legally at fault they don't go around feeling guilty about the incident
We often tear down buildings where mass shooting occur. Not sure about train cars; I think we generally just clean them up and put them back into service after they run someone over. Cars? Well, I'd love to be able to "run a carfax" and tell how many peds this particular jalopy's mowed down before climbing in. But I'd feel ever better if they shredded people-killers after downloading their innards and inspecting the relevant equipment.
Not.
Robot car kills human.
That's the take home for the family.
-- Tigger warning: This post may contain tiggers! --
Not only was this car speeding, but it did not recognise a road side hazard and drive by cautiously.
The police said the woman came "out of the shadows", it's likely that the car couldn't see the women before she stepped out, any more than a human could have (or did, since there was a human driver in the car too who said she had no idea anyone was there to step out).
Most self driving cars DO respond to anomalies by the side of the road and slow down or move over... but again, they have to be things that can actually be detected.
Also, 38MPH on a speedometer is within the margin of error of measurement that it was probably going the speed limit. All speedometers are set to read a bit high, in any case 35 vs 38MPH would not have made a difference to the pedestrian or ability to stop (and some people are saying the speed limit is actually 45, which makes way more sense with a divided median)..
"There is more worth loving than we have strength to love." - Brian Jay Stanley
In a microsecond, it will request the files of each person from Facebook, Google and the IRS and calculate the value of each person' life. If the one person is more important than the other four, those people will be dead before the car even hits them.
#DeleteFacebook
I hope Google's car and others are better than Uber's efforts. If a deer jumped in the road in front of your self-driving car, I guess in Uber's case, it would just plow through it. Aside from any injuries, that could mean costly repairs for the sensors that absorb the impact and to re-certify, calibrate the vehicle.
Humans can adjust to changing situations, they can also ready body language. Most people slow down when they see someone on the side of tge road looking like they are going to step out. An AI cant read that sort of thing. They can only react tl basic things presented to them.
My impression is that they detect and react to the actual physical posture and motion. But they can't read the person and tell if he appears drunk, high, mentally challenged or in some other way odd and likely to do odd things. It's a bit like the difference between a dog on a leash and a street dog with no leash, to a human they pose very different risks. But without programming in a ton of "human" logic they'll look just the same to a computer.
Live today, because you never know what tomorrow brings
An AI cant read that sort of thing.
First of all, self driving cars are not run by an AI.
Most people slow down when they see someone on the side of tge road looking like they are going to step out.
A self driving car can judge such things. And the cars I was involved in do!
The problem imho is that some idiot companies in the USA wanted to reinvent "self driving" technology instead of either partnering or buying european know how.
We have self driving cars on the roads since a decade or longer. Of course only a few still doing their required 100 miles test run (or how many miles that are), of course under supervision of humans (usually a crew and not a single "ersatz driver")
There is no single published incident involving a selv driving car.
And now two companies in the US who are "new to that business" already have several ... yeah, I know "autopilot" is not "self driving", but it is pretty close.
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
You are overcomplicating it. If a car ends up in a complicated situation where it has to guess at how many will get killed, the answer is always to just brake. Get the amount of energy in the collision down, and who knows, some people just might survive. If not, too bad.
Squirrels don't count for the evaluation, you are obliged to not risk anything to avoid a squirrel.
Finally! A year of moderation! Ready for 2019?
Sure...only 3mph, but my Cruise Control can do better than that.
Plus, any reasonably good driver is always scanning the sides to see what might be coming out of driveways or the side of the road.. This time it was a person. What if it's a car next time.
Can the AI spot a car traveling at 50 MPH approaching a 4 way stop and not slowing down? Happened to me just last weekend.
When Fascism comes to America, it will call itself Anti-Fascism, and tell you to give up your guns.
Those are valid ethical questions that have to be answered by the AI and the programmers. If a car is forced to make a choice between killing a squirrel or killing a child, will it treat them the same?
This is not a "valid ethical question". It is just silly.
How will it choose if it has to decide between killing 4 people or 1?
Unlike most humans, the SDC will do the right thing.
But these rare corner cases are not that interesting, because they are ... rare. Far more common are accidents where the correct course of action is obvious: hit the brakes. And SDCs are FAR better at that. A typical human takes about 1.5 seconds to realize what is happening, move a foot to the brake, and start depressing it. An SDC can do it in less than 10 milliseconds. At 70mph, a car travels more than 150 feet in 1.5 seconds. The response time will be even worse if the human is not paying attention.
The car makes no choices.
If there is a potential to crash into someone/something it breaks. And usually it breaks long before that, so that the situation does not even escalate to such a point (decission)..
What the fuck is ethical in avoiding hitting 4 people who cause an accident and hitting a bystander instead? You have some mental problems I think.
What is next? 4 sick old retirees versus a young pregnant mother?
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
An AI cant read that sort of thing.
Why not?
Without the video itself we won't be able to tell for sure. However, from the streetview imagery, there's a number of places she could have emerged from that would've kept her her hidden behind bushes or trees. And I don't think she stopped at the side of the road at all. If she did, she would have had enough time to see the car that clearly was not slowing down for her.
But I'd feel ever better if they shredded people-killers ... they should be smelted, with just 1C degree above their smelting point to make their death painful and s_l_o_w!
Tz tz tz
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
Another key takeaway is that this scenario can now be analyzed and applied to millions of future situations. I just wish all the various autonomous driving companies were sharing their work.
How is that different from the used cars available for sale now? Do you know if any of them ran over someone? Does it matter in that case if there was a human driver or not? It's still just a car.
What a coincidence that it just happens to be a SDC that just happens to kill a pedestrian!
Uh, yeah. 15 pedestrians are killed every day by human drivers in the US alone. What kind of simpleton freaks out about one pedestrian killed in the entire history of self-driving cars?
The car will never have to make any such decision.
What you're so poorly trying to Express is called the Trolly Problem. The answer in this case is that until laws are passed which gives different value to different lives, including that of the occupants, the car should minimize damage as much as possible, and limit the POTENTIAL victims based on simple numbers.
Dog should be avoided (as should all property damage)
A cow should be avoided (as should all property damage)
Cars hitting bugs is generally acceptable (I've never gotten in trouble for my windshield)
These aren't very difficult.
More interesting is should it protect the driver? Do drivers protect themselves (perhaps swerving onto a sidewalk of people to avoid getting T boned)?
I'd think breaking in an attampted to minimize inact damage is a good enough response when coupled with faster reaction time, the trickey part will be for human drivers to learn that they need a bigger following distance.
Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
How will it choose if it has to decide between killing 4 people or 1?
Roughly 3000 people in the US are killed each month from cars, or 36000 a year. Many times that are injured. Less than 10% of these accidents are due to mechanical failure, so in a perfect world we could save 32,400. While I doubt that we'll achieve a perfect world, anything lower than 32,400 is an improvement. Thus delaying an improvement over the status quo due to hypothetical and unlikely events may kill people by delaying progress. Stop killing people :-)
What reason did the Uber car have for going 38 in a 35 zone?
Surely the speed limit was lowered from 45 to 35 for a reason, probably for safety reasons.
Can the car not read road signs? It doesn't have the excuse of "I was watching the road, not my speedo" for a minor speeding offence. Did Uber fail to update the map data when the speed limits changed?
The risk of death being hit by a car below 30mph is relatively low. It increases rapidly as speed increases.
9% chance of death at 30mph.
50% chance of death at 40mph.
Starts reaching 100% fatal over 50mph.
There's a reasonable chance the woman, who may well have been in the wrong, would still be alive if the car was traveling at or below the 35mph limit.
source: https://nacto.org/docs/usdg/re...
There's another study that showed a reduction in speed by 5km/h would result in 30% fewer deaths. That happens to be how much the Uber car was over the limit.
http://humantransport.org/side...
No one is programming a car that way.
The first rule is to anticipate and slow down before anything could happen.
The second rule is to brake.
And the third is to stay on your lane. Except you have a spare lane going same direction.
Neither a programmer nor a car is deciding if it hits 2 3 4 or 1 person. If the thing in its lane is not going away, and the car has not stopped in front of it: it is hit. As simpel as that.
What is next is a realtime auction between the life insurance companies of the potential victims to determine who gets hit.
Run by AI bidding bots?
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
I've seen Christine, I know how this is going to end.
Fully licensed blockchain psychiatrist
It will, but each time, the software will be improved to nearly eliminate the chance of it happening again. Over time, through iterative improvements, driving software will be near perfect.
Meanwhile, no matter how many laws you make, humans will ALWAYS: eat, text, do makeup, yell at their children, look down while searching for something, drink and drive, and engage in every other manner of distraction or impairment that cause accidents. You will NEVER be able to fix this without technology.
AI cars will kill people in ways that humans won't because but car AI "thinks" and operates differently than the human brain. But ultimately it will kill far less people in total than humans, and that's what's ultimately the most important.
Dogs should be targeted.
Added bonus: being able to eat whatever's left on the roadway.
;)
8====D ~~~
I wouldn't have felt guilty at all. 100% fault on the cyclist.
I would be upset she couldn't pay to fix my car though. That should come out of her remaining belongings or cash supply.
What are you talking about? Most humans make the right decision in trolley problems.
Tire markes would mean that you braked to hard locking the tires, causing you to lose traction. A system breaking at optimal efficiency would and should not leave tire marks under any circumstance.
Feminists be like: Go blame the victim, why don't you?
Were there any tire marks indicating the car tried to swerve or brake to even try to miss the pedestrian? Or did it just roll right over her like she wasn't even there?
The last time I hit a deer, it came out of the shadows at the side of the road, and stepped out just in front of me. I didn't mean to hit it, but it got inside my reaction time. I didn't leave swerve marks (well not much), I rolled right over the poor thing like it wasn't there. I don't hate deer. It wasn't callousness or indifference. I was just not fast enough to do anything.
A human driver, valuing the life of a fellow human being (or any other living creature for that matter) would at least try to not hit them, even if that meant swerving into a stationary object (parked car, lamppost, etc) or having to brake so hard they were struck from behind by another vehicle. Did any of these things happen?
It's not that the machine didn't care, but a human would. There was a licensed human driver sitting in the driver seat. They weren't fast enough to do any of these things either.
Or did it just digitally shrug and keep going?
No, you can see that the car is still there in the various photos.
The answer to these questions matters greatly. These so-called 'self driving cars', ...supposedly has a reaction time better than a human driver; sure doesn't look like it from here.
I have no trouble believing that a machine can react faster and often more reliably than me. It cannot however react in zero amount of time. We don't know how to do that. I am running wetware that has a million years of field testing behind it, and yet I am still not perfect. I do not expect the first generation self driving cars be infallible when I am not myself.
Given time they may become the safest drivers on the planet, but not yet, not right now. Besides, even if it reacted in some vanishingly small fraction of a second, it takes time and distance to bring a ton of rolling steel to a full stop.
Is usually around 5mph. It's difficult to keep a car at a rock solid 35mph, even for a computer. Changes in elevation can quickly alter your speed and religiously adjusting for it isn't even always the safest thing to do.
One of the hard lessons I had when driving is that if you slow down too much aggressive or stupid drivers will take that as a signal to go. My first accident was a t-bone where a girl hit me because she was trying to do a left into a busy road. I saw her start to move and put on my breaks. She saw me coming and did the same, but then saw me breaking and decided this somehow meant I was going to come to a complete stop in the middle of a busy street (the only option that would have stopped the accident by then). If I had not breaked she wouldn't have gone and the accident wouldn't have happened.
What I'm saying is there's such a thing as too much caution. Now, maybe if we can get the meatbags off the road that won't be true anymore.
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
"acceptable" - yes, code understands this word
"kill" - equip it with an ECG to detect when brain activity stops, since that's the closest we have to a definition of alive->dead
"dog" - if you're lucky the code is able to categorize obstacles with a tag called ProbablyLivingThing
When you program a robot to make a PB&J sandwich you only use the word "spread" in an instruction after you spend a week defining "spread" using only code that the pile of sensors and motors DO understand.
These cars were told (1) anything that isn't XYZ is errant conditions (2) 99% of errant conditions are a one-step flowchart that says "Brake, beep at operator."
Aaand it's at +5, Interesting. Guess you're right.
Francis :).
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
> if the car has to hit someone
Yeah, no, here's your entire tree of navel-gazing philosophy scenarios you watch the code handle: DetectedObstacle{True|False}
The scenario-handling flowchart (which is mostly a bunch of arrows all pointing at ActivateBrakes) will start at the first sign of anomaly. There's no anomaly evaluation, Christ.
supposedly, watching in all directions all the time and supposedly has a reaction time better than a human driver; sure doesn't look like it from here[/quote]
You're fucking blind then; nice agenda you have there bucko.
15 pedestrians are killed every day by human drivers in the US alone.
So what? We're not talking about 'human drivers' we're talking about your Savoir the Self Driving Car that's supposed to be so goddamned wonderful, and it just killed a human being. That's what's on trial here. I don't give a flying fuck about how many people are killed by human operated cars every day. You fanbois keep trotting out how fucking safe SDCs are supposed to be, yet here's a case where with all the hundreds of millions of dollars invested in developing them, all the media hype, all the cheerleading by fanbois like you, touting how much SAFER it's going to be, yet right out of the gate here's someone dead. Your SDCs don't have real AI, they're 'pseudo-intelligent' at best, the approach to this is all wrong, the technology is being RUSHED and PUSHED on us, and this is the predictable result: someone is dead because of it. Won't be the last person dead because of it, either. I just wonder how many people are going to have to die before someone in authority says "enough is enough" and puts a stop to it.
Will SDCs someday be a helpful tool for humans? Maybe. But we need actual Artificial Intelligence, not the 'pseudo-intelligence' they keep trotting out and CALLING "AI". It needs to be able to actually think, like we do and be aware like we are. That day is not today.
Human drivers cover more miles in a single day than SDCs have covered in their entire history.
The fact that someone was killed in an incident involving the car is sufficient. Many of you are talking about the tech as though it's a pet or has sentience - it doesn't. You can't make a human comparison. This doesn't bode well for autonomous cars dealing with mixed traffic, either. The world is not a perfectly flat, smooth, and dry grid consisting of perfectly predictable, logic-based variables like a testing facility is. Tools that kill people performing their intended function are tools that will likely not be adopted by society. There will probably never be fully autonomous cars, deal with it.
The police chief is using non-technical, human terms, rather than quibbling over semantics of idioms.
"Shadows" can simply mean "obscured from view," and is a common American English idiom.
It could also mean "shadows cast by the headlights" or "shadows cast by the LIDAR beam".
If you're in the shadows of a car's headlights, it's a sure bet the driver can't see you.
-- Sometimes you have to turn the lights off in order to see.
Unlike most humans, the SDC will do the right thing.
The point of the trolley problem is that there is no correct answer. That is how it was formulated by experts in the field. Claiming that there is a "right thing" comes off as cringeworthy levels of ignorance.
I've seen enough road fails youtube videos to have quite a bit of anecdotal evidence that people do not slow down and react extremely poorly when the person does step out.
Also the police chief, who has seen the video, made it clear that the person "stepped out of the shadows" and "would have been difficult to avoid this collision in any kind of mode (autonomous or human-driven) "
Seems like people are hell bent on the belief that autonomous cars can't be better than the amazingly faulty human.
How far do these cars look ahead? In defensive driving, they teach you to look WAY up the road. 150 yards back from the intersection, you are more likely to see people running onto the road than 5 yards from the intersection; it may just be a flash of them seen between vehicles up ahead. Are these cars properly watching as they pull up? They should have to submit high definition video from the moment the car starts to when it stops, from the perspective of a driver. If that person that it hits becomes visible at any time and the AI doesn't show any reaction in some way, "she ran out in front of the car" isn't good enough if you're only paying attention 20 feet before the intersection. I know some people don't notice these things, but a lot of people do and it prevents accidents. I would rather have autonomous cars be modeled after defensive driving techniques and I am concerned that they are not.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
Shut up, faggot.
Actually, most self driving cars don't simply brake. I was in my Tesla S in the leftmost lane of a highway - doing about 65 mph - with auto steer enabled. A car drifted into the lane from the right. The Tesla did brake but it also swerved onto the shoulder to avoid the other car.
This all happened so fast I didn't really have time to react until after the fact. ( It's sort of an interesting question if it knew there was a shoulder to drive onto or if it would have driven me into a ditch if there wasn't one. )
doesn't it?
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
But if that deer was in the middle of a wide open road at night, would you have missed it?
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
The next generation of self-driving cars will be able to run a hundred different braking and steering simulations in a fraction of a second. There is never going to be a 'HAS to hit someone' situation in real life.
-=This sig has nothing to do with my comment. Move along now=-
We often tear down buildings where mass shooting occur.
If a mass shooting occured, then chances are the walls and ceilings have been penetrated by bullets, and that building may no longer be safe to use --- E.g. hidden damage, potential threats to the structure, electrical, etc, could have rendered the place unsafe.
If you disobey the jaywalking laws and a reasonable driver hits you, you are objectively the perpetrator of the accident. What is so hard for you to understand about that? If I'm driving 45mph down street and a pedestrian decides to run across the street, odds are good that I won't be able to break in time. People like you think that'd make me the bad guy if I hit them, despite it being likely impossible between the sheer Physics and normal driver reaction time.
I see this shit all the time where pedestrians will simply disregard the law AND Physics like oh yeah, that big SUV can just stop on a dime if I keep walking despite them having the right of way.
If anyone is the [insert proper medical term defining asshole who doesn't care] it's the pedestrians who smugly walk out in front of vehicles expecting the forces of nature to bend to them.
Clippy: "It looks like you've hit a cow. Would you like help writing BBQ invitations?"
"How will it choose if it has to decide between killing 4 people or 1?"
How would you?
'Shadows' could also mean "Uber already payed me a shitload of money to downplay the situation".
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
If it is clearly the woman's fault, then produce the video for us all to see. Please blur the impact though. I just want to see for myself how much time before the woman entered the lane until impact. Simple, where is the video?
From what I can gather after looking at the intersection and what the police said, a homeless woman walked out from the dark of an underpass late at night into oncoming traffic. The nearest light is a ways from the overpass and it likely would not have been very well lit.
Tempe had clear weather at the time of the accident and it never has enough humidity to have fog or similar things, given that it's a desert.
So what? We're not talking about 'human drivers' we're talking about your Savoir the Self Driving Car that's supposed to be so goddamned wonderful, and it just killed a human being. That's what's on trial here. I don't give a flying fuck about how many people are killed by human operated cars every day.
I appreciate your honesty, but don't worry, we could already tell. It's obvious that you don't care about how many people are killed, and you don't care about whether driverless cars are safer; all you really care about is finding an opportunity to rant and rave and generally act like a dick.
The rest of us DO actually care about how many people die, and we care about reducing that number. This is why you'll never be part of any actual discussion on the subject.
Human drivers cover more miles in a single day than SDCs have covered in their entire history.
So, in other words, driverless cars are 15 times safer?
My brain is magic. It's such a common (and strong) reaction whenever anyone mentions automation or AI that it almost seems like some kind of instinct.
Maybe this has all happened before....
Are you a relative of this woman or something? I'm just reading through the thread and my first impression was that you were a corporate sock puppet, maybe for some Uber competitor, but the hysteria in this post seems more like someone emotionally involved.
The first rule is to anticipate and slow down before anything could happen.
The second rule is to brake.
And the third is to stay on your lane. Except you have a spare lane going same direction.
And the world is rarely that clean. Random shit happens and you have to make a decision. You're on ice, your brakes don't work. Do you ram into a group of people in your lane, or do you hit the parked car? A child darts out into the road in front of you, close neough that you're not going to stop. Do you hit the kid, or go into the ditch?
Vehicles are 1500+kg of mass and momentum, they can't stop on a dime. At some point something will happen forcing the driver (be it human or computer) to make a least bad decision.
...si hoc legere nimium eruditionis habes...
If not guilty the question will ALWAYS be, could the accident been avoided IF a real person had been driving? Could the accident been prevented or if not lessened if the driver tried differently to avoid the death. The machine and it's AI has no value for life. It makes decisions base on logic that no one understands or knows other than a few programmers. Do they decide to be careful and brake when unnecessary 'just in case' or do they program the AI to drive like an 18 year old who has his daddy's hot rod for the weekend?
Until there are standards and rules defined as to what is acceptable AI behavior and wht is expected to be programmed AND that expection is in writting by companies as to their liability should they fail meeting those requiremens, we should all be avoiding this disaster. Most of you have already forgotten Volkeswagon's diesel gate which IS STILL being investigated. Companies cannot be trusted.
I've hit a deer in the middle of a wide open road, at night, while I was going 60 km/h. Something coming at you from the side can be VERY hard to avoid, particularly if it jumps out from cover after your headlights have swept over.
No you're not picturing it. You're in a road, and there is a deer crossing the opposing traffic lane; you would see something like that.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
How will it choose if it has to decide between killing 4 people or 1?
I assume it can just follow right of way laws.If 4 people that jump out in the middle of the road, and the car has the choice of hitting them, or 2 pedestrians lawfully on the sidewalk, or kill the driver by plowing into a concrete post, it should hit the 4 pedestrians in the road.
It's possible the car may misinterpret who has the right of way, but it should be the basis of the decisions of what to hit.
Less than 10% of these accidents are due to mechanical failure
This research indicates mechanical problems are about 2% of accidents.
Enigma
We're not talking about 'human drivers' we're talking about your Savoir the Self Driving Car that's supposed to be so goddamned wonderful,
I'm super confused by this talking point that keeps coming up in the comments. Obviously current prototypes aren't safe; they're prototypes! This isn't a "no true Scotsman" argument; Uber clearly doesn't believe their product is production-ready as they had a human driver in the car. The fact that this is being treated any different legally from a car where the human driver was in full control is absurd.
Once we develop self-driving cars, they will be very safe. We clearly have not done so yet.
The two pedestrians on the sidewalk have the strongest right-of-way. By law the driver has the least, which is a sucks-to-be-him deal. When you get into and drive a car you assume risks that cannot be reassigned to anybody else who is not driving a car, even if they are out in the street in your 'right of way.'
You should put that on a bumper sticker. I bet there are cyclists with cinder blocks in your neck of the woods.
I would imagine second or third generation self-driving cars will be networked together: transmitting obstacle information down the line to cars farther behind, around corners, etc. Having that capability will be inevitable in my opinion.
I can't see government passing up the opportunity to place "road safety cameras" along every street, ostensibly transmitting off-road pedestrian and obstacle movement details to the approaching cars, thereby preventing accidents just like the one that started this discussion. The resulting constant and pervasive population monitoring would just be a pleasant side effect for the new overlords.
Wow, it occurs to me that Facebook would probably put the cameras up themselves, free of charge, just so they could know even more about everyone. Fuck the future, man.
When the only tool you have is a claw hammer every problem starts to look like the back of someone's skull.
In the same way that you could be a paid shill for taxi drivers, I guess you have a point. If you want to see a conspiracy no one can stop you.
Fuck it, just destroy his character anyway. It's the American thing to do. In fact, threaten any children he might have while you're at it.
"Our two-party system is like a bowl of shit looking at itself in a mirror." - Lewis Black
Not to mention haunting. Hail fire, the universal cleanser.
When the only tool you have is a claw hammer every problem starts to look like the back of someone's skull.
Indeed. Good safety engineering practices dictate to always remove system energy as fast as possible when you have no clear strategy. The examples given in the press of these "ethical dilemmas" are bogus. The car in question will not have enough information about the situation and can only a) bake hard without steering (as that could make matter much worse) or only minimal steering and b) tell any following cars what it is about to do. If there is a crash, it can c) directly alert emergency services. This is far, far better than what a human driver could do after his slow brain works out what is happening. Starting to brake that 0.5...1 second earlier an autonomous car will be able to alone will safe countless lives, and that is the comparison with a non-distracted competent driver.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Ah there he is. I knew we'd hear from our resident Cab Company Shill.
"Oh my God. This is terrible. This is the end of my Presidency. I'm fucked."; ~ Donald J. Trump
It makes for stories that sell well, because they anthropomorphize the car. "See, _cars_ will make _ethical_ decisions!" It does not make for any kind useful description of technology as it is complete horse manure.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
The car will go straight on as that gives it maximum brake efficiency. Seriously, this is solved.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Given that there is a lot of distracted driving, unsafe driving, drunk and tired driving in these accidents, at the very least self-driving cars will provide a significant reduction, probably a massive one. Yes, they will still kill people (most technologies do), but far fewer of them than human-controlled vehicles. Control (often craved by humans) does not come with the skills to use that control competently.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Yes, I simplified a bit. ...
But it would not pick a small group of people for not hitting the big one
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
Can a self driving car tell if 2 people on the side of the road are drunk and wobbling all over the place, or if its 2 friends horsing around. Or as another poster said a stray dor or a dog on a leash. A human can see these things and knkw they pose different risks to the driver.
A self driving car knows if the road is iced.
Show me a driver who goes for the ditch when he has to make that decision, and then invite him to the programmers and teach them ho to make the car that decision.
You and other guys here are talking about stuff that is fully out of scope for a normal human but expect other normal humans, the programmers, to blow some artificial ethics into a stupid self driving algorithm?
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
I would imagine second or third generation self-driving cars will be networked together: transmitting obstacle information down the line to cars farther behind, around corners, etc. Having that capability will be inevitable in my opinion.
Off the shelf cars right now are already networked to inform each other about traffic and accidents an other shit.
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
Do you have to work hard at being that obtuse, or does it just come naturally to you?
You fanbois keep trotting out how fucking safe SDCs are supposed to be, yet here's a case where with all the hundreds of millions of dollars invested in developing them, all the media hype, all the cheerleading by fanbois like you, touting how much SAFER it's going to be, yet right out of the gate here's someone dead.
Did you realistically think that self-driving cars would always avoid every fatality? Have you seen anyone make that claim, and can you cite it? Do you believe that the fact that a computer is controlling the vehicle makes the vehicle immune to the laws of physics, so that it can stop immediately? Is the coefficient of friction between the tires and the road any different if the vehicle is operating autonomously?
Secondly - "right out of the gate?" Self-driving vehicles have been on public roads in the US since 2009. They have driven in the neighborhood of 10 million miles across several different cities during all conditions. That is not "right out of the gate". And, in all of that time, 1 fatality has been produced as the result of a self-driving car hitting a pedestrian. Contrast that with the fact that in 2015 in the US, a car killed a pedestrian on average every 1.6 hours. And if you're going to try to gloss over those facts in pursuit of your weird agenda, then you're the fanboi, buddy.
Won't be the last person dead because of it, either.
No shit. Name any risky major venture that mankind has pursued over decades without anyone dying from it. When the Apollo 1 astronauts died on the launchpad, should we have immediately suspended all space operations until we were 100% certain that no one could possibly die from any possible cause? You can die choking on food, does that mean you're going to stick to a liquid diet for the rest of your life?
It needs to be able to actually think, like we do and be aware like we are.
I don't know if you're aware of this, but a lot of people are pretty shitty drivers.
"Our two-party system is like a bowl of shit looking at itself in a mirror." - Lewis Black
Is that what happened here? The road was "wide open", and the person was "in the middle" of it?
"Our two-party system is like a bowl of shit looking at itself in a mirror." - Lewis Black
Wow. We thought you were here to discuss the topic. And here you are, tipping your hand. You're just here for the snark.
I have to agree, fault has little to do here. The fact is the sensors failed to pick out a hazard. I highly doubt that the car goes into super sensitive mode when approaching a known crosswalk. If it were a kid chasing a ball people would not have the same reaction. They'd say the car needs to do better.
However there's still a lot to know. Mostly where the car did start to react, if at all. Talking about lighting and shadows to me makes little sense.
The car was driving 38 in a max 35 mph zone. The car was speeding, so Uber is at fault.
I worked with a lady who killed herself and her child because a cat ran in front of her and she swerved to avoid it causing her to lose control and roll the car. The person in the other lane saw it because they ran over the cat. The irony was overwhelming. I'm pretty sure the SD car would have motored on.
Wow. We thought you were here to discuss the topic.
I'm here to discuss the topic with anyone who is interested in having a serious discussion. I'm not here to discuss the topic with a jackass who is predisposed to demonizing driverless cars and openly admits that he doesn't care about safety statistics.
If it had been a white person that ran over a poor immigrant carrying trash bags you dam well know that that person would have been totally accountable for the actions of themselves , the maintenance of their vehicle and for not properly anticipating the likely actions of poor people carrying trash bags on a bicycle. Visibility and darkness be damned. However since this is a robot and there are billions of dollars on the table for self driving vehicles, the police will naturally take a holistic look at the situation and give the self driving vehicle the benefit of the doubt that they would never ever give a white or black person (assuming it was not a millionaire or city council member driving)
But ultimately it will kill far less people in total than humans, and that's what's ultimately the most important.
You really have no way of knowing that before it happens. Whether humans will be able to string together sensors and 'AI' to be a successful driver.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
How will it choose if it has to decide between killing 4 people or 1?
Oh, that's easy!
It kills the four-person group and then backs up to take out the single person too so as not to be accused of racial, cultural, ideological, gender, wealth, or political discrimination and/or bias by Progressives.
Easy-peasy!
Strat
Progressivism (aka US 'Liberalism'): Ideas so good they need a police/surveillance-state to enforce.
"Moir added that "'it is dangerous to cross roadways in the evening hour when well-illuminated, managed crosswalks are available.'"
Note that it is dangerous to cross a street at night under any conditions. Generally, by the time a driver sees a person in their headlights, there is not enough time to avoid hitting them. That is why you should always use crosswalks at night, (and, as a matter of safety, even during the day.) At a crosswalk, I still wait for the cars to stop even if I have the signal. If there is no crosswalk, I always make sure that I have enough time cross even if the driver does not see me, I would rather be wrong than dead right.
Can a self driving car tell if 2 people on the side of the road are drunk and wobbling all over the place, or if its 2 friends horsing around.
Sure. In either case, the pattern of movement would be different from 2 sober people purposely walking parallel to traffic, and the prudent thing to do is slow down. An ANN should have little problem learning those patterns. Most likely, this is already a solved problem, or considering the millions of miles driven, there would be more than zero avoidable pedestrian deaths by now.
Or as another poster said a stray dor or a dog on a leash.
That also seems like a relatively easy pattern for an ANN to learn.
Do you think the engineers designing these systems are stupid? They have libraries of millions of scenarios, both simulated and real recorded events, which they use for training and testing. It is unlikely that you are going to think of anything new that isn't already handled.
Alternatively, if the SDV is owned by a politically-connected 'liberal' Democrat, it continues directly on it's way and kills whomever is in it's path, returns home, and alerts the liberal Democrat owner to "wipe it, like with a cloth".
Self driving car doesn't mean it will be able to subvert the laws of physics.
When I see a squirrel in the road i stay the course. All but twice they've moved out of the way at the last second.
Scott
I have a hard time believing if a human was driving at the time they would not see her. I am sure the person in the car did see the girl and thought the car would stop if she stepped into the street.
If you go into the ditch your insurance is going to hang you out to dry. The right answer is to plow the kid and then sue the parents for damage to your car.
Read, humans agree with humans in the trolley problem.
I doubt most courts could give a fair and impartial trial--with so much money at stake.
https://www.youtube.com/c/BrendaEM
Hahahaha, just NO.
No one is programming a car that way.
The first rule is to anticipate and slow down before anything could happen.
The second rule is to brake.
And the third is to stay on your lane. Except you have a spare lane going same direction.
"I'm sorry, Dave. I'm afraid I can't do that."
You might want to have the batteries replaced in your sense of humor.
Socialism: a lie told by totalitarians and believed by fools.
anybody know why the AIs all suddenly started to accelerate and steer?
sounds acceleration is prudent and profitable
Good, poor people are cheap to replace.
If there is a potential to crash into someone/something it breaks.
It breaks down, the wheels fall off....? What? How does it break?
No sig today...
Quickly making a sponge cake to lessen the impact damage?
( ÍÂ ÍoeÊ- ÍÂ) orly?
How much do you suppose Uber had to pay out to get the boss cop to exonerate their robot? Bet there'll be a lotta lawyers and coppers shopping at the Tesla dealership next week...
You might find this interesting... it poses an ethical question assuming self driving cars will be able to analyse an accident scenario and decide among various outcomes.. down to including insurance costs etc.
http://www.metronews.ca/news/v...
The article above is informative but if you prefer, skip to the trailer for the "game" ( it was presented at Sundance ):
https://vimeo.com/150956968
Humans should be nowhere near the car traveling at 70 mph, and vice versa.
I do not believe in karma. "Funny"=-6. Do good and forbid evil. Yours, Oft-Offtopic Flamebaiting Troll.
Breaking and swerving simyltaneously sometimes lead to additional probability of sliding
I do not believe in karma. "Funny"=-6. Do good and forbid evil. Yours, Oft-Offtopic Flamebaiting Troll.
You are intentionally comparing uncomparable data ? Or you don't understand elementary statistics ?
In 2015, over 5,000 pedestrians were killed by collisions with cars in the U.S.
3 trillion miles driven by human drivers in the US in 2016. Suppose it is slightly lower or roughly the same for 2015 - that is one pedestrian killed every 600 million miles in 2015.
Autonomous vehicles : 1 pedestrian killed per 10 million miles driven. Autonomous vehicles are clearly 60 times worse. If you opine that data is insufficient until autonomous vehicles complete a trillion miles - I would not disagree except to say that you don't get to opine autonomous vehicles are any safer than human driven ones.
Most of those 10 million miles have been supervised by a professional human, and most of the 3 trillion miles by human drivers have been unsupervised by a(nother) human. So autonomous vehicles themselves deserve much lower credit for the safety their 60 times worse records show.
Bingo Dictionary - Pragmatist, n. A myopic idealist.
Why do we assume that our cities should be places for cars and motor transport? Why do we assume that it's okay to drive at lethal speeds routinely? Frankly, even if you drive safely, you are still contributing to the hidden massacre caused by pollution (in the UK that's about 40,000 deaths per annum, or 10 times the numbers killed by motor traffic).
Still, it's easier to blame pedestrians, so let's stick with that rather than asking the difficult questions of why we think its okay for so many people to die every year.
Neither a programmer nor a car is deciding if it hits 2 3 4 or 1 person.
Interestingly neither are humans. Anyone who thinks that humans attempt to make a rational choice during an emergency situation has never been in the situation before. People at best slam on the break and try to avoid hitting something. Rarely if ever are they even able to register if that something is a child or a deer.
I remember slamming on my breaks one day to avoid a wallaby which jumped out from behind a car. My wonderful passenger armed with all the hindsight in the world smugly said that it was illegal to swerve avoid wildlife. Reality: The fact it was wildlife didn't even register until the car was nearly stopped.
The problem imho is that some idiot companies in the USA wanted to reinvent "self driving" technology instead of either partnering or buying european know how.
There is plenty of American know how to partner with or buy, but Tesla and Uber thinks that they know better than those with experience.
Well, from what I've heard Uber will shut down their self-driving project so there is always that.
Truth, but I will never meet someone like your mamma!
That sounds more like an error, than planned action.
Tesla autopilot isn't designed for that kind of avoidance. It's a driving aid that keeps you in a lane and at a constant speed. It can only change lanes when the driver tells it to.
More likely the other car obscured its view of the road markings, and it suddenly thought it was way off the centre line for the lane and moved over. The only emergency action autopilot is designed to take is braking.
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
What if it kills something we eat like a cow?
This happened to a friend of mine.
He was driving in a rented car through Brazil, and hit a cow. Stopping was poor advice, as the inhabitants of the village surrounded him. Some of them had guns, and were angry. He was Canadian and did not speak Portuguese. He was explaining the theory of insurance and police reports of accidents required by the rental. When a gun was put into a place that it would hit him if discharged, he offered US$50; we used to carry US$ because it was cheaper than CDN$ to change. They treated him like a long-lost son afterwards (he could have had any of their daughters). However, he got out of the place as quickly as he could in the car - luckily it could drive. He discovered that the car rental company took his word on the wreck, and just charged the repair costs to his AmEx.
Typically the police say "it would be inappropriate to comment on an active investigation".
How much money was this doofus paid by Uber to issue this uselsss wishy-washy statement. It's only purpose is to keep the Uber share price buoyant.
""Shadows" can simply mean "obscured from view,"" - no, it can't. I've never heard anybody use "shadows" to mean "obscured from view". They are two different things. You could be sunbathing in direct sunlight behind a wall and be "obscured from view", it has nothing to do with "shadows" whatsoever.
Can a self driving car tell if 2 people on the side of the road are drunk and wobbling all over the place, or if its 2 friends horsing around.
Sure. In either case, the pattern of movement would be different from 2 sober people purposely walking parallel to traffic, and the prudent thing to do is slow down. An ANN should have little problem learning those patterns. Most likely, this is already a solved problem, or considering the millions of miles driven, there would be more than zero avoidable pedestrian deaths by now.
Or as another poster said a stray dor or a dog on a leash.
That also seems like a relatively easy pattern for an ANN to learn.
Do you think the engineers designing these systems are stupid? They have libraries of millions of scenarios, both simulated and real recorded events, which they use for training and testing. It is unlikely that you are going to think of anything new that isn't already handled.
I think he is questioning the state of current AI technology rather than trying to insult you and AI engineers. If all these scenarios were already 'handled' and considering that this car can effectively see in the dark with it's LiDAR and night vision cameras, you'd think that this Uber cab would have spotted that woman, identified the threat she posed, slowed down and then stopped in response rather than running her over. Clearly something is not quite as advanced and 'handled' in the world of self driving cars as you'd have us believe.
hit the brakes. And SDCs are FAR better at that
Really ? So why do humans manage to kill only one human every 600 million miles driven ? And the "Artificial Intelligence" unleashed upon various cities managed to kill 1 human within 10 million miles driven - many of them overseen by yet another human to override the stupid AI if necessary ?
The "rare corner cases" are not interesting to you because you don't understand them.
Bingo Dictionary - Pragmatist, n. A myopic idealist.
The National Transportation Safety Board usually takes a minimum of a year to complete an investigation with as many complexities and contributing factors as this one; they must have put in unbelievable effort to get the investigation done and the report written in less than 36 hours.
What kind of simpleton freaks out about one pedestrian killed in the entire history of self-driving cars?
Human drivers : 30 trillion miles driven in last decade, 50,000 pedestrians killed.
Self-driving cars : 10 million miles driven in last decade, 1 pedestrian killed.
Humans 60 times better than machines restrained by humans from killing pedestrians.
Bingo Dictionary - Pragmatist, n. A myopic idealist.
Sad story. We learn in driving school to only break and not swerve ... but even then you can be partly guilty when one crashes into you from the rear.
Reminds me about a friend of mine who came late to a party ... brought by an ambulance. He had a collision on his motorbike with a wild boar. The bike was a complete loss and he was pretty bruised.
Next morning a score of hunters went out to find the boar and shoot it (assuming it would be heavily wounded and would be suffering) ... they found lots of blood traces and skin but never the animal ... it seems it also only got a bit bruised.
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
By pushing the breaks ...
If that is incorrect english, correct it, instead of trying to make a smart comment. After all your comment looks pretty retarded to me.
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
Understandable, I actually never had an encounter with wildlife on the road.
Only saw them on the meadows or in the trees besides the road.
The only nearly accident was a bird of prey that nearly crashed into my windshield. No idea what it saw there and was trying to catch. But with far over 100km/h on the highway that probably could have ended very nasty for me.
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
Oh whatever. I put those shits in their place all the time. Come at me with a cinder block and you're going in the dirt, bloody. Cyclists are subhuman vermin. I cheer every time one gets killed.
I looked up "break" in the dictionary and found a verb.
break brÄk/ verb
1. separate or cause to separate into pieces as a result of a blow, shock, or strain. "the rope broke with a loud snap"
2. interrupt (a continuity, sequence, or course). "the new government broke the pattern of growth"
https://encrypted.google.com/s...
How do I push one?
No sig today...
Exactly. There is no super deep ethical questions here. Itâ(TM)s just greed and rushing to market with no government oversight.
These self driving systems should be more human than human. Why was this car not outfitted with IR detection to see the women in the road REGARDLESS of lighting!
Okay. The word you were looking for is "brakes". English has a lot of homonyms, which are differentiated in writing by spelling...
"I do not agree with what you say, but I will defend to the death your right to say it"
No, kind of like a gun kills its victims.
That's why it's imperative that we ban humans.
And this is relevant to what the OP said, or the article, how?
You push the brake with your foot.
Some people call it "slam" though.
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
Thank you :D
The smart ass is still harassing me with stupid comments, like "look what I found in the dictionary about 'break'" instead of pointing out, I made a typo.
I hate stupid people that think they are smart.
Cost free eBook I read (by iBook/Kobo/Amazon/ObookO/Gutenberg etc.): "The Green Odyssey" by Philip Jose Farmer.
Hate to be that person, but a typo is when you type it wrong, not when you don't know about the rule.
A computer does NOT have to be programmed on what is acceptable to kill just like drivers don't have to think that way. Furthermore, the word "kill" is inherently inflammatory in this context. A car is not programmed to kill and your claim is absurd on its face.
It is not acceptable to "kill" a dog or a cow nor will a car even detect a spider. A car will avoid collisions always but cannot avoid them when put in a no-win situation, just like human drivers.
Ok so if I promise you a bridge, will you give me $10,000 now? Because that's basically what the "self driving will save lives" promise is.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
Then a reduction needs to be demonstrated. They aren't doing that by hitting pedestrians so soon in the game.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
It's probable that these cars are just not even looking to the sides in that way. They likely do not track all pedestrians close enough to be a problem like a human driver, and just assume that anything not right in front of them can be ignored for the most part. I don't think image recognition is really at the state where a car can have a live understanding of its surroundings in this way. And predictive pathing sounds like it is probably hard enough that they just didn't want to spend the effort.
A human can see another walking in some direction and deduce what their intentions are. A car is not going to see the worn down grass path indicating a crossing point. It is not going to use predictive psychology if its sensors are turned in that direction.
It is likely closer to: if(At this time there are no obstacles in front of me) Drive
Troll is not a replacement for I disagree.
That's not really the point though. A computer is not a human. They have a chance to program the choice in a computer, unlike a human. So what are they programming?
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
Unless the squirrel and the child are the same size, the answer to this is pretty straightforward. Furthermore, a squirrel is less predictable. Behavior and size will dictate this every time regardless of the value you place on the respective lives. Teach your child not to run in front of cars, problem solved.
Regarding the 4 lives vs 1, it is unlikely that an AI would encounter this hypothetical in a context where its programming would be able to make a difference, however one could argue that it could "choose" to "kill" all 5 and that would be acceptable since that car iwould not be the cause of the accident in any case.
The responsibility to avoid these situations lies with the people who cause them, not with the computers that can't solve them. It's not the train that's to blame when a person jumps in front of it.
Somehow this girl managed to be totally invisible to the car. That doesn't happen in a road; sure people get partially obscured by other vehicles and traffic controls and the like but they have been designed not to obscure vision. It is important to know why the car didn't see the girl.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
This is a great way to monetize! Buy the Super Life Value subscription so Uber cars weigh your life more favorably!
Rich people rarely have any skills, so they're much lower on the priority list.
#DeleteFacebook
Murderer!
If the one person is more important than the other four...
The problem where you have to decide whether to let a train run straight and kill 5 people or switch the track so that it only kills 1 becomes more complicated when that one person is the president and the 5 people are his bodyguards. My guess is that passenger safety outranks jaywalker safety, but I'll bet a lot of situations are covered by "panic and brake".
He's getting rather old, but he's a good mouse.
Squirrels are fine. You could blow a tire on a bunny.
He's getting rather old, but he's a good mouse.
This should be assessed as if a human was driving the car.
It’s profoundly unwise to attempt for a company to bribe the police in the US. Given the NTSB is involved, it would be exponentially worse of an idea.
The NTSB’s standard faire is investigating airline accidents, and are used to far higher stakes than a mere traffic collision.
-- Sometimes you have to turn the lights off in order to see.
Those are valid ethical questions
I wouldn't say that. At least not in the near term. Currently, the programming or "AI" we are talking about can't distinguish between a living thing and an inanimate object. We've even seen cases where they have difficulty distinguishing between a large white object and empty space (or perhaps a close vs. distant object). If it sees something in its path, it will stop. If something enters its path in less than its current stopping distance, it should brake and attempt to steer away from the object if there is a clear path. If there is no clear path, it should just attempt to stop.
So, what about in the future when the machine is smart enough to distinguish between a person, a dog, a trash can, churches, lead, a duck, whatever? It might not be an issue at that point. With that much intelligence in the machine, it should have much better predictive ability and much greater ability to process larger amounts of incoming sensor data. Meaning it can have more and/or better sensors that can see farther outside the vehicle's path and process the data more accurately to estimate whether something outside its path is likely to move into the path, which could necessitate decreasing speed in case it has to stop, or changing lanes if it seems likely (perhaps even slightly) that a vehicle/pedestrian/whatever will move into the SDC's current lane.
Or it could be that he is saying, "Had a human been driving, they'd have hit her too."
So what are they programming?
Stop in case of obstacle. Change lanes if clear. That is what they have been programming from the start and that is precisely the best outcome. The other possible scenarios are littered with a stupendous amount of variables that make programming them impossible.
Point is that this isn't a case of ethics, it's a case of best normal response. If that response involves killing a family with 5 children while the elderly person in the car survives in order to claim that there was an ethical problem you need to first prove that the decision wasn't the best course of action which is impossible to do given the variables involved.
I've seen this framed as a "protect the driver" vs "protect the pedestrians" but the reality is the former simply isn't possible. Any action that puts the former in threat involves a whole lot of variables that are impossible to predict: e.g. do you swerve to avoid one child at the expense of the driver in the car and how can the car tell the difference between hitting a parked car as a result (the protect the pedestrians route) vs hitting a car full of children (same scenario as protect the driver but with a now even worse outcome than just running over the single child in the first place).
What we program is what we can control: Stop. Change lanes (on your side of the road only) if its safe to do so. Its exactly how existing collision avoidance systems have been programmed thus far and these have already saved an incredible number of lives.
How much is the AI insensitivized to hit the squirrels though?
You're entirely right. This is probably a net safety win.
My concern with it, though, has always been that a human knows that a dog standing on the sidewalk can become a dog in the road in 2 seconds and it should be given a wider clearance and slow down a bit.
A garbage can on the curb on a windy day is just as much of a potential hazard as a garbage can in the road.
And a human might have known that there was no visibility if a biker with plastic bags decides to step into the road.
I think overall, driverless cars are going to be a huge boon to safety, especially at night in the suburbs and rural areas where bars are far away and taxis are limited. But this is a hard problem, and we shouldn't stop at "good enough".
Version 1 won't. Version 12, post "the s or driving massacre at _______" definitely will.
* Self-driving cars are required to have three green flashing lights, front and both sides (LED green)
* Newly produced human driven cars are required to have purple flashing lights, same configuration (LED purple)
* By 2025, everyone will be driven nuts by the flashing lights, and we'll all figure out how to ride bikes while wearing pajamas on our way to the broom factories.
"There is no god but allah" - well, they got it half right.
Probably because humans are allowed to drive without IR vision. And it's just one more system you'd have to calibrate that could screw up.
But she walked out in front of t he car. IR won't help that.
Given 210 million licensed drivers and 13,500 average miles per driver, and 37,500 traffic deaths per year, you end up wityh something like 75 million miles per death, not 600.
One death in 10 million miles is probably reasonable given some statistical variance.
But you never had any math classes beyond high school algebra, so alarmism.
and when the autonomous car has a bug or has been hacked to speed up and steer into people....what then?
With your limited brain capacity, try to get simpler data .
Pedestrians killed (remember the topic being discussed) : over 5000 per year. https://www.cdc.gov/motorvehic...
Miles driven : over 3 trillion per year. https://fred.stlouisfed.org/se...
Demand back your money from your "algebra" classes. Because they admitted ineligible students like you who failed arithmetic.
Bingo Dictionary - Pragmatist, n. A myopic idealist.
Now you're just being silly.
No, when you have every opportunity to make the choice before it happens then it IS a case of ethics. Perhaps the AI is too simplistic for that kind of reasoning, in which case it is too early to have automated cars on public roads.
Laws are rules for the court, but merely a bottom bar to hit for life. Think beyond laws in your actions always.
When walking, riding a bike, and riding a motorcycle I always try to make eye contact with drivers coming up on me that could cause me problems. I think having someone behind the wheel who isn't in control of the vehicle is particularly dangerous for this reason. I can look someone in the eye to make sure they see me, and assume that means I am OK to do something when the vehicle really has no idea.
I really think that cars need a clear indication to traffic when they are being controlled automatically verse when a human is actually steering.
And when you go crazy and start eating your own toes for some bizarre sexual gratification, then what?
Why do we assume that our cities should be places for cars and motor transport?
Because they are designed that way (at least in North America)
Please stop APK.. you're only hurting yourself.
Well, this _is_ /. so it fits.
Many people were expecting this to happen sooner or later.
Thanks to nearly a century of lobbying, the USA has a pedestrian-hostile culture and many states have car-centric laws (eg, jaywalking) which state that pedestrians must give way to cars except at designated crossing points and even then can only cross on a green light.
Growing up with these kinds of rules translate into assumptions that people generally don't walk ontot he road unexpectedly, which translate into robotic rules that people NEVER do that - meaning they're not setup to "expect the unexpected"
A human driver should have seen the pedestrian on the median some distance off and reacted (slowed down or changed lanes) even if she was stationary because someone standing there is likely to move. The robot just kept going thanks to cultural assumptions programmed into it that became rigid operating rules.
(Anticipating kids running onto the road is one of the classic tests for danger perception in many countries. Virtually all US drivers fail such tests badly, which is why trading your USA license in for any EU one generally involves at least a full license test.)
Sure, there shouldn't be a pedestrian on the freeway (or a deer either) but I'm not going to run him over. I sure as hell don't want my robot car to do so. Nor do I want it to ignore the 12 point stag, or cow on the road - one of those would come through the windscreen and kill the vehicle occupants. Ditto for the unexpected solo road cone. There might be a pothole behind it.
Arizona has to be one of the worst possible places to test self-driving vehicles. It's extremely car-centric, shitty and hostile for pedestrians and the number of edge/corner cases that happen is low, meaning the human supervisors get complacent. My bet is that the twit in the driver seat didn't even have eyes outside the cabin until after the poor woman in question was bouncing down the road.
The next generation of self-driving cars will be able to run a hundred different braking and steering simulations in a fraction of a second. There is never going to be a 'HAS to hit someone' situation in real life.
Unless self-driving cars are going to be capable of breaking the laws of physics, there will always be one 'HAS to hit someone' situation in real life--that of somebody stepping into the road immediately in front of the car, in a situation where it cannot brake fast enough and cannot switch lanes.
Poor people are often more expensive to allow living.
It will avoid a collision uf it can. It can't know the outcome of the actions of others. For example, an incoming car is suddenly in it's lane. It will brake to reduce the force of the impact. If an adjacent Kane is empty, it may seek to get out of the way.
Sorry, thought we were talking about fatalities, the relevant metric.
Ok, your example and concern is still meaningless. The woman was in a situation where a human driver would have also hit her.
But keep up the alarmist tone. Won't someone think of the edge cases!
But sometimes they don't, then watch out.
http://lifeisaroad.com/stories...
They likely do not track all pedestrians close enough to be a problem like a human driver, and just assume that anything not right in front of them can be ignored for the most part.
As a runner and cyclist, I can assure you that most human drivers don't bother tracking pedestrians.
The woman was in a situation where a human driver would have also hit her.
Ok, run your time machine, replace the car with a human driver at the same place and same time. Let us see.
If your time machine is broken, shut up about what would have happened and concentrate on what happened. And learn some math.
Bingo Dictionary - Pragmatist, n. A myopic idealist.
Its because of the law of averages. If you kill 1 person every 600 million miles SOMEWHERE in the distance you kill a person. If you stop counting after 10 million miles then you're not allowing the car to travel the other 590 million miles. In addition you need to take into account the average distance for that particular location or style of road and the driving conditions. Let make this a fair comparison before we shoot all the machines.
my problem was I slowed down. e.g. speed limit was 45 and I dropped to 35. If I'd kept an even speed no harm or foul.
Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
Pedestrians are willfully ignorant of the law and violate it under a potentially fatal misunderstanding of physics. In general, across the US the only place where pedestrians have the right of way over cars is at crosswalks at controlled intersections when signally indicates that a pedestrian may pass the cross walk. At all other places where it is legal for a pedestrian to cross a road they are almost always required to yield the right of way to traffic on the roadway. Another common law is that pedestrians are not allowed to cross a roadway between two controlled intersections where crosswalks exist.
70% of pedestrian fatalities do not occur at crosswalks. 34% of pedestrian fatalities involve a pedestrian with a BAC .08 or greater.
"Lack of speed can be overcome. In the worst case by patience." --Znork
Ooh, we got a real tough guy here.
Or if you're living under a government that's figured out how to leverage social media, you can look for all that info in just one place.
fluffernutter:
But if that deer was in the middle of a wide open road at night, would you have missed it?
If I had seen it, I would have *attempted* to miss it. Where I live every adult has a when-I-hit-the-deer story. Sometimes people joke about putting tape over their deer whistles and going out to get them some venison, but really everybody knows it's a bad idea to hit them. You can pick up a thousand dollars worth of damage in just the time it takes to move your foot from one pedal to the other.
The point I was trying to make is that it does not look (at this point) like the car demonstrated what in humans would be called "Depraved Indifference", (which is what I think the poster ahead of me was concerned about,) it looks like it failed an impossible test.
There is of course a lot that we don't know yet. The N.T.S.B. report won't be out for a while yet, none of us have seen the dash cam video from the car, we don't know what the cars black box says, we don't have testimony from the human backup driver, we don't know if the bike with all the bags may have hidden the human being behind it, we don't know if she waited until it was right on top of her and then stepped out, etc, etc.
I do expect that some very bright engineers are going to go over the events with great care, and if it is possible to improve their algorithms I'm sure they will. Of course the engineers may be up against the limits of what physics permits, in which case the car will only be as good as an alert and attentive human would be.
If a ball comes into the street from behind a parked car, I assume there's a kid following, and react accordingly (probably slamming the brakes). AIs can do the same. If a person just comes out from behind a large vehicle at the exact wrong time, without any warning, I'm going to hit that person.
"When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
fluffernutter :
It is important to know why the car didn't see the girl.
It may have seen her, but assumed she would stay on the sidewalk, not jump out in front of it.
I fail to see the moral quandary in that particular dilemma, at least with the current President.
The AI doesn't have to be programmed with any fancy criteria. It can brake as hard as possible. It can swerve if safe, if not not it can just brake.
"When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
You must be new here.
"When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
My father-in-law was hit by a deer once while driving to work.
"When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
Self-driving cars, when widely deployed, will save lives. (Otherwise, they won't be widely deployed.)
"When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
Same as when the nuclear power station has a bug and blows up: Boom. Then, if the industry is smart, it fixes that bug in future instances. (The nuclear industry is too greedy for that in many instances.) Also, system like these come with a lot of redundancy, so "one bug" will never do anything like that.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
at least with the current President
Well, yeah. I meant in principle. I'd be just fine with DJT stepping in front of a bus if it wasn't for Pence.
It can swerve if safe...
That's a big "if" with a TBD definition of "safe".
He's getting rather old, but he's a good mouse.
Not all of us with reduced or absent empathy are psychopaths. That isn't even an APA diagnosis anymore, it is a pejorative. Antisocial personality disorder, the preferred term, requires multiple criteria, and diminished affect is just one.
refactor the law, its bloated, confusing and unmaintainable.
1. But what if the machine "driver" kill one pedestrian every 1 million miles, but the human "driving assistant" stops them initial 9 times and lets a pedestrian be killed the 10th time?
2. If data is not enough, one can't say that "machines are better drivers" either.
3. Nobody is "killing" all machines.
Bingo Dictionary - Pragmatist, n. A myopic idealist.
There are a number of problems with that reasoning:
1. You can't really draw statistical conclusions from a one off occurrence. The fact that one pedestrian was killed after 10 million miles does not mean that driverless cars kill 1 per 10 million miles. It could be 0.01 per 10 million, or it could be 10 per 10 million. We won't really know until we have a larger sample size.
2. It doesn't account for types of miles. AFAIK the majority of testing for driverless cars has occurred in city settings, while the majority of miles driven by humans are on highways. One of those environments is far more risky than the other.
3. It only looks at pedestrians and not at collisions as a whole. If the number of fatal crashes as a whole is significantly lower then driverless cars may still be safer, even if they do turn out to cause more pedestrian deaths.
Still, at least you're actually talking about the numbers, which is a big improvement over the OP. I'm fully willing to accept that driverless cars could currently be less safe than human drivers, but I'm not going to accept that conclusion based on the emotional ravings of some lunatic. Show me the numbers!
1. You can't really draw statistical conclusions from a one off occurrence.
That is fine*. But I see people here drawing conclusions that self-driving vehicles are far safer than human driven ones, from nothing other than their own imagination, or far less data, or misleading data like yourself. And I also see you not correcting most of the ones I noticed.
Your own conclusion is highly suspect. 15 pedestrians killed per day is not comparable at all to "entire history of self-driving cars" for anyone with a clue in statistics.
2. It doesn't account for types of miles
Yup, most self-driven car miles have been overseen by humans - so if they avoided accidents it is completely unknown how much credit can be given to the humans.
3. It only looks at pedestrians and not at collisions as a whole
In many places, self-driving cars are prohibited from carrying passengers** . Why would you ask for evidence that legally and likely cannot exist in large quantities if you have a clue about statistics ?
Show me the numbers!
That is a big problem. With the available data you have made misleading statements, what would you do with more numbers ?
The raw data is "proprietary". These vehicles would drive around, potentially killing people. The data we have is for miles driven with human oversight, potentially sanitized to remove their culpability.
This is absolutely worth freaking out. The burden of proof is on these companies about to profit from firing human drivers. One important strategy for nurturing a conspiracy theory is to invert the burden of proof. You are doing that nurturing, not the critics of self-driving vehicles here, like it may seem to shallow readers.
* : I mentioned as much in one of my posts where it was more directly relevant. I don't mind mentioning it again, except to conserve a semblance of readability by not flooding a post with disclaimers.
** : Even though the "driver" or "driving assistant" in this case looked more like a passenger.
Bingo Dictionary - Pragmatist, n. A myopic idealist.
A good rule of thumb is never to swerve to avoid an animal. This should be the default. The risk to human life is too great, generally. The unpredictable always occurs. In individual situations, there may be enough time or sufficiently low risk that swerving reduces all damage. But the bias for not swerving should be so strong that leaving your lane abruptly should be a final choice, not an early or first choice. When you and those you share the road with finally die of old age, there will be enough gratitude among everyone to compensate for the odd blown tire.
A good rule of thumb is never to swerve to avoid an animal.
There's a world of difference between a squirrel and a moose. If there was room, I'd absolutely swerve to avoid a moose.
He's getting rather old, but he's a good mouse.
We can define "safe" conservatively here. If the car can determine that swerving would be safe, it can swerve.
"When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
I'm not sure "safe" is as simple as you're trying to make it. A person steps out in front of you. Is it safe to swerve if it means hitting a dog? A curb? Is it safe to swerve into the oncoming lane if there's no traffic there? Maybe you're defining "safe" as not hitting anything or violating any road regulations, but a slightly more liberal definition could save lives.
He's getting rather old, but he's a good mouse.
This is a BS ethnical question: If the driver (human or AI) has time to ponder ethnics then it has time to avoid both. Thereâ(TM)s a reason the speed limit in a parking lot isnâ(TM)t 65 mph.
I'm thinking of "known to be safe", which presumably means if the area swerved into is going to be free of obstacles for a moment. That would rule out hitting the dog. Other definitions are possible, but that's the one I think is likely to be applied.
"When you have eliminated the unacceptable, whatever is left, however improbable, must be the truthiness" - Holmes
I can see you thought out a well reasoned and logical retort. To respond to your definition of "a BS ethical question" imagine if you will....
A self driving vehicle is driving on a highway with a posted speed limit sign of 65 mph. The vehicle is tracking and monitoring all traffic and pedestrians around it. A child on a bicycle darts out 10 feet in front of the vehicle. Next to the vehicle is another vehicle with an elderly person in it. An example of an ethical question in this case would be "do you swerve into the vehicle carrying the elderly person or run over the child?" The self-driving car has broken no laws at this point, but simple physics would prevent it from avoiding injuring or possibly killing one or the other.
And yes, parking lots don't have a speed limit of 65mph, but highways do. Once you've progressed past the point of sitting on daddy's knee to drive around the parking lot you'll learn these things.
So you are saying that when you sit in the drivers seat of a car, you have license to run over anybody in your way?
Humans can adjust to changing situations, they can also ready body language. Most people slow down when they see someone on the side of tge road looking like they are going to step out. An AI cant read that sort of thing. They can only react tl basic things presented to them.
This is absolutely the kind of thing that AI can do today. You just need enough data to train a classifier for pedestrian propensity to walk into the road. With enough AI miles driven, there will be enough data. Google and Tesla almost certainly already have enough data for that. Though in this case the problem was likely visibility, so it's not really relevant.
The CAR failed to see the pedestrian the car is 100% at fault if it was running in auto-drive mode. The driver was doing just what an auto-driveing car would have done NOT pay attention to the road the car is doing the driving and that's what the automaker advertised auto-driving abilities..the car failed badly. cops an idiot go back to writing tickets to people who don't stop at stop signs..or get some tech company to do that for ya too.
Jack of all trades,master of none