Slashdot Mirror
Blind Man Test Drives Google's Autonomous Car
Velcroman1 writes "'This is some of the best driving I've ever done,' Steve Mahan said the other day. Mahan was behind the wheel of a Toyota Prius tooling the small California town of Morgan Hill in late January, a routine trip to pick up the dry cleaning and drop by the Taco Bell drive-in for a snack. He also happens to be 95 percent blind. Mahan, head of the Santa Clara Valley Blind Center, 'drove' along a specially programmed route thanks to Google's autonomous driving technology. Google announced the self-driving car project in 2010. It relies upon laser range finders, radar sensors, and video cameras to navigate the road ahead, in order to make driving safer, more enjoyable and more efficient — and clearly more accessible. In a Wednesday afternoon post on Google+, the company noted that it has hundreds of thousands of miles of testing under its belt, letting the company feel confident enough in the system to put Mahan behind the wheel."
Grow up. They have done 200,000 miles with a person sat in the driver's seat to ensure he can take control if anything went wrong. On a pre-programmed route this is a very stable system and he had someone beside him in the passenger seat (I also wouldn't be surprised if it was dual-control so the passenger has access to a brake pedal). Meanwhile this technology could eventually change the lives of millions upon millions of disabled people, damn right it deserves the publicity. With your attitude we'd never have wheelchairs or crutches or surgery, all things which, the first time out, could have resulted in injury but have been life changing tech for millions.
Boy, if that's not one of the most appropriate metaphors for our time...
Soon, they'll just jack us into our pods, and grow us for the power we generate. :-)
"Flyin' in just a sweet place,
Never been known to fail..."
In the UK you are not allowed to drive unless your eye-sight meets a minimum standard. Is it legal for a 95% blind man to drive in the USA?
For christ's sake, stop talking about the google car! Every time it's mentioned, anywhere, it pushes its release to two years in the future.
Later she was called to the police station to make a statement. The police had arrested the driver. He said he had not seen the crossing because there was thick fog (mildly overcast). Then they discovered that he was registered partially sighted. He had cataracts.
He was convicted of:
His comment to my wife at the police station? "You've spoiled my day". He simply did not realise how serious his offense was.
So I applaud what Google is doing, because I've worked with computers for nearly 35 years, and human beings for over 40, and if the system is designed I would trust the computer over the human being any day of the week, and double on Sundays (drunks with hangovers).
From scarped cliff or quarried stone she cries "A thousand types are gone, I care for nothing, no not one."
"'This is some of the best driving I've ever done,' Steve Mahan said the other day.
I guess he usually uses those pavement reflector thingies to drive by braille.
Driving home tonight there was a young kid playing quite near the road, so I dropped my speed in anticipation of him doing something stupid. He didn't, but I did wonder about the google car making those sorts of calls. I'm sure these google guys are pretty clever and have thought of all these things... are there any video's of self drive cars reacting to these sort of situations?
Like that feeling you get when you see someone else on or near the road and you aren't completely sure that they have seen you and you react by lowering your speed to avoid a potential collision. It's got me out of trouble a few times. If there was an accident you probably wouldn't be at fault, but you've gone one better and seen the accident coming and avoided it.
I'd want to see lots of video evidence of a self drive car doing this sort of thing before I'd be happy sharing the road with one.
Yes, because Google (and the authorities letting these cars on the roads) would have *never* thought of the possibility of pedestrians running in front of these cars.
Quick! Get in touch with them and bring this to their attention!
Most likely an autonomous car can react quicker to an obstacle running in front of it faster than a human can.
And given the average human's driving ability, it probably fares no worse when it comes to being in the correct lane at complex junctions.
Maybe it will need two more orders of magnitude of testing and refinement before it can be included in cars that the blind person can be alone in, but progress is progress, and this is surely a milestone?
Of course I will jailbreak my car when it comes with such technology, so that I can add my own AI modules, such as "HunterKillerMod" that turns the car into a pedestrian killing machine. And "DestroyAllCyclists" too, obviously (who won't have that installed?).
You should read about what the system is actually capable of.
And to go a little further, technology doesn't get sleepy. Technology doesn't get distracted by cell phones, GPS systems, or the radio. Technology won't have a blind spot. This is going to be an incredible advance. I'm much less worried about a driverless car hitting a pedestrian than I am the average driver hitting one.
I'm a leaf on the wind. Watch how I soar.
The trouble with this type of driverless car tech is that it's going to be as brittle as the AI it's based on. It may work fine for normal, complex or not, situations, but the day a child runs out in front of it in a way it's not been programmed to handle there's going to be a tragedy.
I'm not sure many human drivers will have driven 200,000 miles without ever ever ever having had even the most minor of scrapes. To me, it sounds like even in the test phase this car is a lot less brittle than the wetware AIs we have driving cars already.
It's one thing if this is going to be used as a glorified cruise control, but another entirely if it's meant to be used without a qualified driver ready to take over whenever appropriate.
Not if the AI in those cars is less brittle than the AI in our brain, even if it is still somewhat brittle.
For this type of tech to be safe in an unrestricted environment (e.g. on public roads) it needs to be backed by human level AI (i.e don't hold your breath), not just an expert system using lasers and cameras to stay on the road and read the speed limit.
Your assertion is equivalent to "if computers are ever going to play chess properly, they'll need a human to stand by, and correct moves for them when they get it wrong". News flash – computers can't play chess perfectly, but they sure can play it better than 99.99999999% of humans.
I asked elsewhere too... are there video's of google cars reacting to this sort of situation?
"not just an expert system using lasers and cameras to stay on the road and read the speed limit"
I hope they implemented an expert system using lasers and cameras to follow the rules of a road.
Actually a child running in front of the car is exactly what the car performs perfectly at avoiding.
What the car can't do is tell the difference between a stationary object and a human.
This is never an issue on a route the car has previously driven on, however the issue is stationary objects on a completely new route where the driver is unable to assist the car to understand the difference between a stationary object and an object.
Either way my point is that the Google car is amazing at anything that is moving, people, cars, lights or even stationary cars. Its really not that far from being perfect and already amazing in areas it has already been.
Did they think of the possibility of driving over a cliff-edge while out of GPS reception?
Or what happens if a bridge collapses? Does the car detect the void underneath it and stop, or just think it's a steep hill and plummet over the edge?
Does it detect ice, snow, oil, sand before the wheels are there? What about fire? What about an accident happening to the tanker in front of you and you ploughing through the spilled petroleum because the car doesn't "see" it? What about kids throwing stones off the top of a bridge onto the passing cars (common problem in the UK - someone died just the other month from this)? Is the car looking UP too and determining their intent?
There are a BILLION and one problems, that only happen once in a lifetime. But if that causes you (OR ANYONE ELSE - sod the blind person, I would complain to the highest authority if a blind person was driving a car around my area, with or without a permit, and risking pedestrians and other driver's lives) to die early, or be at raised risk of injury, there's a lot more things to consider than you can EVER detect with sensors OR ever account for in programming and testing.
This is why even a jumbo jet - so of the most highly automated and tested machines in the world - has TWO HUMAN OPERATORS. And even there, they have TWO because the first can't be trusted on their own (proven by that recent thing with the pilot).
If you honestly, seriously, think that you can reliably determine the outcome of a machine complex enough to obtain all that data, you're an idiot. You *CAN* verify a system like an airbag control, or ABS, because it's isolated and has the tiniest amount of actual code running the thing that you can (and DO) mathematically verify.
You can't verify a system on this scale. It's like trying to verify a Kinect. You just cannot guarantee what it will detect something as just by a simple test of something similar. And this is orders-of-magnitude more complex, more important and more deadly than a stupid games console.
When was the last time a standard wheelchair did 80mph when the user pressed a button/pedal? When was the last time a crutch was fitted with ABS to help it stop in time because it went so fast?
There's progress, and there's fecking ignorance of the scale of the problem.
They can also drive safely millimetres (like inches but smaller) apart from each other, massively increasing the capacity of the existing road network.
I've seen that thing MERGE WITH MOTORWAY (freeway right?) TRAFFIC!!! 8@~~
It's bonkers clever. I want one. Where we all just sit around the table inside it having breakfast.
You have an unrealistically high opinion of the skill with which the average human drives a car. American drivers for example manage to kill roughly 40,000 people a year, that's a little over a hundred a day.
IAAR (I Am A Roboticist) and this car can handle random stuff in the environment. It has fairly simple algorithms doing object detection, sophisticated software figuring out where the road is and whether any objects are in the road, and a massive, massive data set of supporting information about the road telling it where all the relevant road stuff is (lights, signs, street markings, all of it). If it's not sure what to do, it stops, gently. The problem isn't random stuff in the environment ("Ooh! I see something! I'd better stop!") it's the fact that it won't run without the massive data set of supporting information about the road. It can only run on that road; that's where the brittleness comes in.
Can you verify yourself?
I know I can't. I've even driven into the back of someone in heavy traffic because a sudden hail storm fogged up the window.
Fact of the matter is, get this system to cope with most situations and put a STOP!!! button on the dash and you'll solve most traffic and traffic accident problems.
Automated cars are also unlikely to rip along at 80-90 kph in a 50 zone like the psycho-cabbie who nearly ran me down on my wake-up walk half an hour ago, too.
I do not fail; I succeed at finding out what does not work.
If you honestly, seriously, think that you can reliably determine the outcome of a machine complex enough to obtain all that data, you're an idiot.
I don't think you can reliably determine the outcome. But that doesn't change when the 'machine' is a human. Let's run objective tests to find out whether the automated vehicle is more safe, less safe or about the same. I believe that's what's going on.
I don't know if there is but from what I understand, this thing is incredibly intelligent. Screw kids running in front of the vehicle, it's capable of anticipating asshole drivers not paying attention and cutting you off. It has sensors in all directions, (like 360 degrees at all times instead of maybe 130 at a time) and watches everything. Wired ran a rather long article about it recently and it was quite the read. Whether this is "human level AI" or not, it is certainly a very clever machine. They aren't "Programming" these cars, they are "driving" it around and letting it learn on it's own.
The problem is technical and social/moral.
We are willing to accept the fact that a human occasionally makes an error. We are, however, not willing to accept when a machine makes an error, not to speak of the occasional errors made by software engineers. That's the social or moral problem. Who's fault is it if there is a software or hardware glitch?
But there is also a technical issue. I personally would not drive in a car programmed by Google engineers, because I am not confident that these people have the experience to develop such high-integrity systems. I want to see the CVs of these engineers first. On how many high-integrity systems have they worked so far? I know plenty of people with an AI background, and trust me, I don't want these to program my car. I'd also need to know which programming languages and development tools they have used, see the source code, and would like to know which formal software and hardware verification methods were used to verify the code.
It would also be a good idea to publish the source code for software used in planes like those of Airbus, but unfortunately they haven't gotten so far yet. However, there are plenty of reasons to have more confidence in airplane safety than in the safety of autonomous vehicles developed by Google. BTW, one reason is that it's technically much easier to fly a plane (without landing, which is still mostly done manually) than to programmatically steer a car or make a robot walk.
Did they think of the possibility of driving over a cliff-edge while out of GPS reception?
If the Internet is to be trusted at all, I'll take the chance of a self-driving car careening off a cliff due to lack of GPS reception over the chance of a human careening off a cliff because of GPS reception.
Does it detect ice, snow, oil, sand before the wheels are there?
Humans certainly don't... but there are already automatic traction control systems that do an excellent job maintaining the vehicle's footing in all but the more extreme situations - I can't imagine it would be that hard to send that data to the pilot AI and have it react by slowing down. Also I'd imagine it would be easier for the computer to detect ice and such using sensor data (IR cameras to detect road surface temp, lasers reveal changes in surface reflective properties, etc.)
What about kids throwing stones off the top of a bridge onto the passing cars (common problem in the UK - someone died just the other month from this)? Is the car looking UP too and determining their intent?
Again, I doubt humans would do much better. The radar systems on an automated car could conceivably be used to detect objects that may hit the car even from above and some evasive/mitigating action could take place - with better reaction times than a human driver.
This is why even a jumbo jet - so of the most highly automated and tested machines in the world - has TWO HUMAN OPERATORS. And even there, they have TWO because the first can't be trusted on their own (proven by that recent thing with the pilot).
Again, though I don't keep careful track of these things, there seems to be more incidents related to human error than automation error. Specifically the humans overriding the automated systems to correct for a problem that didn't actually exist.
If you honestly, seriously, think that you can reliably determine the outcome of a machine complex enough to obtain all that data, you're an idiot.
Humans are essentially machines much more complex than that, and have tens of thousands of years worth of historical precedent for doing incredibly stupid things despite having accurate information - yet somehow they are more trustworthy than a machine just by virtue of not being a machine? This kind of argument instantly refutes itself.
How do you test the system for these things? Tens of thousands of hours of real-world driving. Considering all a human needs to legally operate a 2-ton projectile is roughly twenty minutes worth of testing (if you're lucky!) I'll take my chances with the machine.
=Smidge=
"Human Level AI"? Humans aren't exactly doing a wonderful job out there on the streets.
I wouldn't be at all surprised if the current autonomous systems are already safer than human drivers. Are they going to be infallible? No. But we need to compare the overall safety rates of computer controlled vehicles vs. human controlled vehicles and see what is better.
Most of the problems you cite are *highly improbable*. Nobody is claiming that a driverless car will never make a mistake, but the facts are many automotive fatalities each year: http://en.wikipedia.org/wiki/List_of_countries_by_traffic-related_death_rate
This is because humans make mistakes. Our current system is not 100% safe. A replacement system does not have to be 100% safe, just better.
To evaluate a driverless systems success it is not meaningful to look at the least likely cases, but to look at overall whether or not it will reduce of increase number of deaths. If 90% of deaths are due to drunk drivers, and driverless cars only fail when say a bridge collapses (a very rare incident), then this will be a net win.
I'd like to know how it can handle hazard road conditions, like Snow, Ice, etc. Also, what if something mechanically goes wrong with the vehicle, how will the A.I. respond to that, or even notice it. If I smell smoke, or something burning, I'll pull over to check it out. Seems like so many variables to consider, I think it will be decades before we have A.I. that can deal with all this.
-- By all means let's be open-minded, but not so open-minded that our brains drop out.
Yes, because Google (and the authorities letting these cars on the roads) would have *never* thought of the possibility of pedestrians running in front of these cars.
+1 Funny
The example was "the day a child runs out in front of it in a way it's not been programmed to handle". What part of "in a way it's not been programmed to handle" did you fail to comprehend?
You can only prepare for situations you think of in advance, while a human has the advantage of being able to judge new situations.
If you think it's possible to prepare for situations that you haven't even thought of, I have a car to sell you...
This type of system is, by its nature, going to be reactionary. For every eventuality you can program it to handle, you can come up with an infinity of more situations that it doesn't handle.
The thing is, there is an infinity of situations like this you can come up with, and no way for the car to be able to handle them all. No matter whether you can say to three or all of the above that "yes, the next version of the car can handle this", we can always think of more situations it cannot handle.
For this to be feasible, the roads have to be dumbed down to the level of what the car can handle. I.e. something like a Sci-Fi railway where only authorized cars are allowed on it.
Meanwhile, in the real world, people like their freedom.
Why would it need to tell the difference between a human and a non-human? If it's not in the way, it's not in the way. If it's moving in a manner that will cause it to be in the way, then react accordingly. A human standing at the curb and suddenly running out into the street is no different, functionally, than an empty trash can getting thrown into the street by a gust of wind. An obstacle is an obstacle regardless of what it's made of and regardless of whether or not it's sentient.
=Smidge=
I know some human drivers who have no idea how to merge onto the freeway.
Frankly, I have much more faith in a Google computer driving my car then I do the other humans on the road.
To your point, the other big reason I'm a fan of driverless cars: Technology doesn't get drunk (see also: Bender B. Rodriguez).
Reaction time is also cut considerably, as is the time it takes to physically perform whatever act is deemed the best course of action. If "slam on the breaks" is the action, the car doesn't have to lift its foot off the pedal and move it over to slam the brake -- the car's already braking.
A child running in front of a car is a recipe for disaster either way, but the kid is probably safer with the driverless.
Was that you? Sorry, I was trying to get FP...
I don't know if there is but from what I understand, this thing is incredibly intelligent.
If it was truly smart it would stay in the garage and never come out. Driving is a curious game, the only way to win is not to play.
This is never an issue on a route the car has previously driven on, however the issue is stationary objects on a completely new route where the driver is unable to assist the car to understand the difference between a stationary object and an object.
I can see cases where that matters - the car needs to swerve to avoid an accident and it can't decide that swerving one way hits something inanimate and swerving the other will hit a human. But in any normal circumstances the fact that there's a stationery obstacle of anything like human size (even small human, even baby) should mean the car either manoeuvres around, picks a new route or stops while the human occupant of the vehicle decides what to do next. As much to protect the car and its occupants as people/cats/etc who get in the way.
Why would it need to tell the difference between a human and a non-human?
No. Just plain no.
The water from a sprinkler at the side of the road is not a danger I need to stop for, even if it registers on a radar. Stopping would cause a danger, not preventing it.
Neither is it a danger when soft branches from a fallen tree that sticks out into the road. Nor do I worry about the leaves blown by a leaf blower. And the crow sitting pecking at the dead squirrel will take off before I get to it. The kid sitting at the side of the road trying to revive a dead squirrel won't.
And what about the policeman standing outside the road, signalling you to stop?
Or the guy in the broken down car in front of you who waves at you out the window to pass?
Expert systems are anything but.
The trouble with this type of driverless car tech is that it's going to be as brittle as the AI it's based on. It may work fine for normal, complex or not, situations, but the day a child runs out in front of it in a way it's not been programmed to handle there's going to be a tragedy.
The question is, how prepared is the average driver? I don't mean hazardous driving but according to stats (1, 2) I found there's about 3 trillion miles driven, 6 million crashes and 40,000 fatalities in the US each year. That's one crash for every 500,000 miles driven - that sounds incredibly little to me but it says 3000 billion miles. Even if you consider that near-accidents and emergencies happen more often they're many thousands of miles apart. Nobody is really able to stay alert that long for something that doesn't happen 99.99% of the time.
It's great what you learned at your driver's exam, but most people aren't there. At best they're simply cruising and need to snap into emergency mode, not counting all those who aren't paying sufficient attention, are tired, fiddling with the stereo, their phone, distracted by passengers, fail to react, panic, do stupid things like swerve into opposing traffic, on alcohol, on drugs and so on. Nor do they have a 360 degree view or IR vision or any of the other tools an automated car would use. By far most often the threat is obvious if observed and the solution usually equally so.
After all, there's only so many things you can observe about the people around you too, you can see the child but you can't know what it's thinking. It's just the basics of that it's a child, what trajectory it's on, is it unaccompanied and in all honesty there's not that much you manage to process while you're driving by at a fair speed. It's mostly attention, if that child runs in direction of the road I have to react instantly and break which will cut a typical response time from 1.5 seconds to 0.2 seconds. It might just not be that relevant to a computer that's still analyzing all possible problems simultaneously.
Live today, because you never know what tomorrow brings
Can the AI fail at very rare circumstances? Yes. Will we see AI failures? Yes. Will we see the AI repeating the same mistake? No. It can be patched. Will we see people driving while drunk in ten years, despite thousands having proven beyond all doubt that it's a really stupid idea? Definitely. AIs can fail, humans fail more. The AI just has to get to less than 1'000'000 deaths per year, and it's already an improvement.
"Humans are essentially machines much more complex than that, and have tens of thousands of years worth of historical precedent for doing incredibly stupid things despite having accurate information - yet somehow they are more trustworthy than a machine just by virtue of not being a machine? "
I believe it is the fact that we (humans) built it, that makes us not trust it. We know how stupid humans are, how could we create anything smarter than ourselves?
Cant wait till this project gets shelved in 3 years for no good reason and everyone needs to learn how to drive again...
It's also an excellent excuse in case something went wrong: "sorry, I couldn't help it: my car caught a virus!"
cpghost at Cordula's Web.
If your crude inbuilt sensor system can smell smoke from the engine, the check-engine light has probably been on for the last two years.
I'll take my chances with the machine.
You do that, but not on any road I drive on.
Yes, humans make mistakes. But those are mistakes. Machines only do what they are programmed to do, and go down in cascades when something unexpected happens. I would rather have an accident caused by human mistake every now and then than a future where all traffic stops and backs up for hours because a branch fell onto the road and the systems raise an alert for "unknown, may not be safe to pass", and don't understand the policeman who waves at you to go around it despite having to cross double yellow lines.
No thanks, you play in your playground as much as you like, but keep it to yourself, please.
Thing is for every problem you point out with a AI driven car you can point out 5 problems with human drivers. Humans frequently mess up in hazardous conditions, especially if they aren't used to them, meanwhile an AI car is going to be programmed for all possible conditions before it will ever be released into the wild. As for something being wrong with the car, that's what sensors are for. We have to rely on imperfect queues like smell, the AI on the other hand should be plugged straight into the onboard computer and have an excellent overview of the car's health. It might miss corner cases but once again, humans will miss many more. Also humans will often suspect something is wrong and carry on anyway because they can't be bothered to check it out, while the AI can be forced to pull over and demand a fix before carrying on.
There will still be deaths on the road if we switch over to 100% AI controlled traffic but I'll be damned if it won't drop the road toll to a tenth or less than what it is now. That's a ton of lives that will be saved, as well as the added convenience of not having to drive yourself. Of course convincing people to give up control to a machine is going to be a tough sell.
Did they think of the possibility of driving over a cliff-edge while out of GPS reception?
Or what happens if a bridge collapses? Does the car detect the void underneath it and stop, or just think it's a steep hill and plummet over the edge?
Google cars use only GPS to get direction. The actual driving is done with laser grids and radars (for near distance) and video camera (for long distance). So the car doesn't drive according to what it "thinks" should be there according to the plan, but it drives according to what it "sees" (with its sensors). Any of the situation you mention will end up with the car detecting a lack of drivable surface, stopping, asking its GPS for an alternative route and going another way.
Wherever a human driver will react the same, or will be too busy getting distracted with on-board enterteinment/smartphone/passengers/news paper, etc. and fall into the hole is left as an exercice to the reader. (yup, we've already had stories on /. of clueless drivers wrecking their cars because the GPS told them to go a certain way).
There are a BILLION and one problems, that only happen once in a lifetime.
And that's why you put the stuff on extensive testing. Already in the range of hundreds of miles on actual raods with actual traffic in the case of Google cars.
Yes, it won't take into account some really weird exceptions. But... Humans make mistakes too, and mostly in normal everyday boring situations (because they are boring and the brain kicks into "autopilot" routine mode). (And chance are that some of the really weird situation are going to be "missed" by the human, not because the human wouldn't have had reacted correctly, but because the human was to bored to pay attention). Also even if you, the human driver, think of hundreds of situation which might be missed by an IA, but where you think you'll be able to react correctly, I can probably think of situations where you drove perfectly well, but still got into an accident because of some other driver.
At some point in time, we will reach the situation where an autonomous car (even including the accidents due to weird rare situations) will cause a lot less casualities than a human driver (who might just not be paying attention).
The point of this publicity stunt is to show that, given the current extensive testing the cars have undergone, this point in time is nearing soon.
And, also, the advantage of autonomous cars is, as the wierd situation happens, they can be analysed and the programming can bu updated, leading to even less casualities on the long term. .
Whereas, with human drivers, you can't just magically "programatically remove from the road" asshole, idiot and dristracted drivers
You can't verify a system on this scale. It's like trying to verify a Kinect. You just cannot guarantee what it will detect something as just by a simple test of something similar. And this is orders-of-magnitude more complex, more important and more deadly than a stupid games console.
You can't prove *mathematically* that the autonomous car will be perfect in absolutely every single situation (juste because there is a potential inifity of such situations). But you can prove *statistically* that the autonomous car is better and safer than a human driver based on the number of accidents and casuality caused by both. And overall this *will* mathematically increase the safety on roads.
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
Speaking of GPS how will it deal with errors on maps? Apparently human beings have enough trouble noticing when their GPS directs their 40 tonne truck down a dirt track next to a perilous cliff.
const int one = 65536; (Silvermoon, Texture.cs)
SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
They can also drive safely millimetres (like inches but smaller) apart from each other, massively increasing the capacity of the existing road network.
Stopping distance doesn't change that much. The reaction time becomes smaller, but the braking time stays the same. It's fine for normal use, but when the car in front collides with an oncoming vehicle or something falling off bridge and comes to an abrupt stop then your driverless car still needs almost as long as a human-operated car to come to a safe stop without hitting the vehicle in front.
I am TheRaven on Soylent News
"Meanwhile this technology could eventually change the lives of millions upon millions of disabled people, ..."
Another tech already did, it's called 'a cab'.
The AI just has to get to less than 1'000'000 deaths per year, and it's already an improvement.
Not necessarily. Not if it means that the average speed of transportation goes way down. Or that the "rare" incidents that defeat the AI causes major traffic blocks, because the expert systems decide to STOP in the face of the unknown.
We have chosen to have a dangerous traffic system. It would be much safer if we limited all cars to do a max of 10 mph. I am certain that almost all the deadly accidents would go away. But we do not want that. We want the freedom and speed, despite the dangers.
If you want to live in a padded room, go ahead. But don't force others to, in the name of safety. Living is dangerous, and that's what makes it worth living.
And "DestroyAllCyclists" too
I believe this is enabled by default if you set the locale to en_US.
I am TheRaven on Soylent News
The drive-up ATM's at Citibank branches in the NY area have had Braille labels on all the buttons for years. Seemed kind of silly up until now, you certainly would not want a blind person standing in the driveway using the ATM, and I certainly hope a person requiring Braille labels on an ATM would not be behind the wheel of a car. Not knowing Braille myself, I always assumed the labels said "Get out of the way!!! You're standing in a driveway!!!".
But now I realize that Citibank was preparing for the eventual release of the autonomous car.
"We make our world significant by the courage of our questions and by the depth of our answers." Carl Sagan
My problem with these carrs is that they won't break the laws when necessary.
Right now every rational person who is not impaired physically is speeding, rolling over stop signs or running over stop signs where they are ridiculous. I am speeding 20-25 mph in my complex when I do not see children around, but if I seen signs of children anywhere close to the road (this happens in the certain time of the day) I slow down to 5mph (the rule here is that one has to assume that at any moment the stupid child will run to the road at its maximal speed). I also maximize the distance between my car and children my crawling on the wrong side of the in-complex road. Autonomous car will continue driving 10 or 15 mph per hour in this situation.
Autonomous car will be like an old Asian lady who got her driver license 3 months ago. The real old asian ladies already have major impact on traffic.
What I want from autonomous cars (any cars for that matter) is to be able to form intelligent ad hoc trains able to drive in a very sync manner on the road, with almost no lag in accelerating or decelerating between front car and back car because they all connected and when the front car of the train had to break, the signal goes immediately at computer stock market trading rate to all the cars in a virtual train. (most importantly this happens when the front car accelerates). Only that soultion will ease the traffic.
If we cannot expand cheaply our roads we need to make them much faster, not 55 mph (speed limit on 270), not 70 mph (actual average speed when traffic is still smooth), but 100 or 120 mph.
I do not believe in karma. "Funny"=-6. Do good and forbid evil. Yours, Oft-Offtopic Flamebaiting Troll.
Why would it need to tell the difference between a human and a non-human?
There are a few situations where this could be important. Consider a cat runs into the road on the right and a child runs into the road from the other side to get the cat out of the road. A human would typically prioritise not hitting the child. If the AI doesn't, and hits the child in preference to the cat then it's not going to look very good. If there's only one obstacle, you want to avoid it. If there are two, then you want to avoid the most valuable ones, and generally we consider humans to be more valuable than anything else you're likely to collide with.
I am TheRaven on Soylent News
The day a child runs out in a way the human behind a steering wheel is not able to react to due to their incompetence, blood alcohol levels, sleepiness, or distraction, there is always a tragedy.
This tragedy has nothing to do with whether a machine or a human is controlling the car. It's a tragedy of an unfortunate circumstance.
It is possible however that on an average the machine does better than or equal to a human. To determine if it is so, it requires testing. Which is being done. So what exactly is the problem? Why do you assume that the human level of intelligence is the end-all and be-all of doing everything? Time and again it has been demonstrated that human intelligence is biased towards certain kinds of tasks. And it is debatable if driving is one of those, considering it is something humans have started doing only in the last 150 years or so at most.
Entia non sunt multiplicanda praeter necessitatem.
As explained by other, the car *does* slow down, and even eventually halt when exposed to situation it thinks it can't handle.
Also, the car has much lower reaction times. So in some situations, it doen't really need to slow down, it will react immediatly if needed, whereas a human driver will need to slow down to make room for slower refelexes.
(The distance between autonomous vs human-driven cars on the motorway, for example).
"Sufficiently advanced satire is indistinguishable from reality." - [Tips: 1DrYakQDKCQ6y52z6QbnkxHXAocMZJE61o ]
Technology won't have a blind spot? I'm not so sure I'd agree with that. Potentially it won't, but implementations aren't perfect especially when trying to bring down the price.
And a human driver uses the superstopbreakpedal thus can avoid the situation better than the computer?
In case anyone is interested, Udacity will be offering CS 373: PROGRAMMING A ROBOTIC CAR again starting in April. The first class is finishing up and Sebastian Thrun does an excellent job demonstrating the concepts required to make the self-driving car happen.
a wheelchair isn't a two ton death machine controlled by computers hurled through the universe.
They're using their grammar skills there.
Wrong, quite a bit of the stopping distance is reaction time. Doesn't mean you can drive millimeters apart but perhaps 30% closer.
I think he's referring to the fact that a digital driver could have as many "eyes" in the form of cameras as it needs, arrayed in whatever way works best. It can have a 360 degree ring of cameras on the top of the car, for example, which has no blind spots at all. (I mean, unless you manage to crawl under the car via a sewer system or something....)
Compare that to a pair of forward facing eyes, with an elaborate system of mirrors to try and allow them to see behind the car as well as in front. Lots more blind spots, and they can only look in one direction at a time.
why can't we have it both ways? the computer can safely handle 90% of the issues but when it detects an "unknown, may not be safe to pass" then the car comes to a stop and has the human take over. remember there is still a human in the car, why not let the human handle the edge cases.
if the car sees it far enough ahead of time it may be possible for it to warn the human of the upcoming hazard and have the human take over before they even reach the obstacle without having to stop. and then once navigated past the obstacle the autopilot could be turned on again.
to be honest most of the time i would trust a computer more than most of the human drivers out there. humans show really bad judgement especially during poor rood conditions such as when there is snow and ice.
...Steve Mahan != Steve Mann.
(Note to Google: a similar test with Steve Mann has the potential to be really, really interesting.)
As the others point out, humans fare way worse than cars on this. If you smell smoke, you know that something is wrong with your car. If a computer is plugged into the standard OBDII port on your car, it can tell you exactly what is wrong by checking an array of sensors before the smoke even starts.
watch the TED talk with Sebastian Thrun - it shows the car making a left turn, stopping for pedestrians in the crosswalk, and continuing.
If you're going to only rely on sensors, your taking a big risk. Sensors can go bad, and in fact it usually ends up fatal in many case, IE - Space shuttle disasters, Plane Crashes, Train Wrecks, etc. I've had sensors go bad on my car many times, when their was nothing actually wrong with the car. You can't compare a few lines of code and some sensors to the Human Brain! Even military drones need a human to make decisions for it. Until we see robots that can think for themselves, self driving cars will be decades away.
-- By all means let's be open-minded, but not so open-minded that our brains drop out.
I actually really like the U.K. proposal to do effectively this.
They want(ed) to link a train of cars to a lead car driven by a professional driver (only on motorways). You'd merge on, connect to the train (somehow, not physically) and bam, done.
I think that's a great tech for children as well.
I read an article a while back about the cars. It turns out, when they programmed them to follow the rules of the road exactly, they couldn't get anywhere because other drivers continuously broke the rules. So they had to reprogram the cars to allow for bending the rules in certain situations. The example that stuck in my mind was at a 4-way stop sign, the car has to inch forward to indicate its intent to pass through the intersection. Otherwise, the other drivers just ignore the car and keep going in turn, despite the rules of the road stating that's illegal.
Sexual intercourse is kicking death in the ass while singing. ~Charles Bukowski
I agree seriously how many bridges does the gp cross that regularly collapse?
the only one i know of during my lifetime that was within 1000 miles of me is the I-35 collapse in MN about 5 years ago? and i still have never been anywhere near that bridge.
More likely it's a step towards ganging the cars together to act as a train on long freeway hauls to gain the benefit of reduced gas consumption through better aerodynamics. The passenger cell of the car would have to be greatly stiffened in that case I think, but good crumple zones plus automated braking upon an accident should help that greatly, and ganging the cars together should be safer than a train in an accident since everyone would be in seat belts and their own safety cell.
Back in 1989 I was on a tour of the IBM factory in Rochester Minnesota. The factory had robotic forklifts and trams that moved parts around and shared the same workspace as the employees. There were also a few forklifts driven by humans. This was pretty advanced stuff in 1989, and the first question someone asked was if there were ever any accidents. The tour guide laughed and said, "It's the ones driven by humans that you have to watch."
Proverbs 21:19
Additionally, once all cars have this system, it won't even need to allow for lag for the driver behind to react, it can broadcast a signal to brake and a whole row of cars can instantly come to a halt, so an end to one car braking and causing a massive pile up. I suppose the car doesn't care what the obstacle is that's suddenly appeared, an obstacle is an obstacle, but watching some of the videos, the system paints various obstacles in either yellow or red, pedestrians red. I assume this is some kind of risk assessment or prioritisation - i.e. if a child runs out in front of the car the system will prefer to scrape the parked car next to it than hit the person. To make this kind of judgement in sub second time in a 360 degree arc is exactly the reason why computers will make better drivers than humans in the future.
... With your attitude we'd never have wheelchairs or crutches or surgery, all things which, the first time out, could have resulted in injury but have been life changing tech for millions.
http://www.ncbi.nlm.nih.gov/pubmed/2383378
Wheelchairs still kill people (or they did, until at least 1987). They are nonetheless life-changing and undenyably a good thing
/* This comment may not be thread-safe */
Compare that to a pair of forward facing eyes, with an elaborate system of mirrors to try and allow them to see behind the car as well as in front. Lots more blind spots, and they can only look in one direction at a time.
And there are lots of drivers incapable (at least for periods) of even managing that. The one huge benefit of this system is that it never gets distracted by something on the radio, or the phone ringing, or wondering what to pick up for dinner, or by the idiot who just cut it up. I wouldn't be surprised if the vast majority of accidents are caused by momentary lapses in judgement (there will be a lot that are caused by plain old bad driving, but over time experience, the legal system, and ultimately crashes, should weed those out).
All the machine has to do is make fewer mistakes than the human. When it gets there and you still take issue, the problem is you.
I'm interested to know what happens when this car is overtaken or undertaken by filtering cyclists and motorcyclists.
I did read some where that there was this case where the gps did not make it clear that you needed to stop and wait for a ferry and some one did not stop and the car went in.
what about GPS having you drive on restricted roads? roads that are gated or cut off but the GPS shows that it goes though or having a truck go down a road that is no trucks or dumb stuff like sending a truck down the tail of the dragon.
Tens of thousands of people die from utterly avoidable road accidents every year, but you'd throw out the technology to prevent that just because it can't anticipate an event that might happen once in a lifetime? You'd happily let tens of thousands die because of the minute risk of one person dying in a scenario the car hadn't been configured to deal with (even if it was then trivial to make that configuration for future uses)? That seems a bit like throwing the baby out with the bath water.
I am sure I am not the genius who came up with this first. And I am sure I heard it somewhere before voicing it for the first time.
I can't find a reference to UK proposal. Can you give me more tags?
I do not believe in karma. "Funny"=-6. Do good and forbid evil. Yours, Oft-Offtopic Flamebaiting Troll.
Most likely an autonomous car can react quicker to an obstacle running in front of it faster than a human can.
With the added bonus that if the computer doesn't react fast enough, it won't drive off and leave the pedestrian bleeding in the road to protect its insurance premiums or hide the fact that it's had a few drinks before setting off.
I don't understand most of these what ifs, you act as if they're unsolvable. My *PHONE* can detect a one-way sign through it's cheap camera and a simple app, these cars will have a wide variety of sensors and more powerful computers. In addition there are standardized signs for no trucks, restricted, do not enter, etc... The car can easily read these signs (OCR could even be employed) and simply route around it. If a car wanted to make the next left, happens across an orange detour sign, or cones in the way, it will instantly recalculate and just route around it. It's really not that hard to imagine these solutions.
why can't we have it both ways? the computer can safely handle 90% of the issues but when it detects an "unknown, may not be safe to pass" then the car comes to a stop and has the human take over. remember there is still a human in the car, why not let the human handle the edge cases.
Because the human, in this case, is blind?
Planes fly themselves too, but still we put able bodied pilots in them.
humans show really bad judgement especially during poor rood conditions such as when there is snow and ice.
Of course they do. But they also have a flexibility that any expert system will lack.
Yes, we can do a lot of things to reduce the number of accidents, but do we really want to? Replacing roads with rails, limiting the top speed to 10 mph, or only giving driver's licenses to women between the age of 28 and 50 would all drastically reduce the number of accidents. That's not necessarily the point.
Simply put, I don't want to see cars on public roads that can't follow the directions of a policeman, or that'd stop at a red light next to a pedestrian slapping his hand with a crowbar.
Expert systems may have the statistics in favour of them, and while the average Joe might not reason as much as we want, the automated car cannot reason. Once these hit the road, I expect a whole new set of problems, including crimes and mischief taking advantage of the certainty that there won't be a reasoning driver in these vehicles.
We know how stupid humans are, how could we create anything smarter than ourselves?
I would think you would build upon the intelligence of hundreds if not thousands of people, much like all of the technology our civilization is based on, or do you happen to know how to manufacture everything you use from scratch including microchips, displays, etc.?
It's not clear whether a sprinkle of water, blown leaves or a crow would have a radar signature large enough to warrant action, so on that point I can't really accept the argument. For exactly that reason I would expect a threshold of obstacle assessment anyway.
Pedestrians standing in the road, such as the police officer and the guy with the disabled vehicle, Do present difficult problems, sure, but not in the context of this thread. The difficulty comes in the form of extra instruction given to the driver that the vehicle might not recognize. In terms of object avoidance, they are still objects to be avoided.
Even for a nuisance stop, it would presumably be as safe a stop as practical.
=Smidge=
Oh, I'm sure there will be a new firmware to enable such behavior soon enough.
Random Thoughts From A Diseased Mind (Not For Dummies)
Generally if the sensor on the car fails it will flag up an issue - you drive it to the garage, they check and find the engine is not the issue and replace the sensor. In other words you have the right amount of failover to ensure that you get your sensors checked but don't end up wrecking your engine without significantly increasing the cost with redundancy systems. If those sensors were handling pedestrian detection you can bet on having a lot more in the way of failover, additional sensors, tests running specifically to determine if the sensors are working, etc. There are also lots of situations where self driving cars would be capable of taking over from humans sooner rather than later - motorway driving for instance, is reasonably predictable, it's also boring and liable to cause drivers to stop paying attention, I'd be happy to see computers driving us on motorways in the near future even if we're decades away from them handling the school run.
" do you happen to know how to manufacture everything you use from scratch including microchips, displays, etc.?"
Nope - but I don't have them running anything important without human intervention either.
I don't know how to build a car either, but I'm the one driving it.
FWIW -- I agree that a computer driven car is going to be far safer than anything operated by a human (or monkey, snail, slug, etc). I was just trying to point out why people may inherently be against it.
The problem is this is just another nail in our coffins. Human are not the best chess players. Humans are not the best Jeopardy players and now humans are not the best drivers. The list is only going to grow until we find there is nothing we do better than computers. This technology will eliminate millions of jobs. It will even change buildings. What will happen when Wal-Mart finds out that it is cheaper to deliver goods than it is to build huge stores? How many square miles of land are now just parking lots that will not be needed anymore? I am glad that I am over 60 years of age because this and other computer advances will eliminate so many jobs that a person born today will find it extremely hard to find a job or a purpose for living.
All the machine has to do is make fewer mistakes than the human. When it gets there and you still take issue, the problem is you.
The quantity of errors is not the issue. The impact is. When it gets there, and it and everybody else in the greater traffic area is delayed for hours, and I take issue, the problem isn't me.
Belief that we want safety at the cost of freedom is why we have abominations like TSA, and why kids grow up without ever having climbed a tree. Life is precious because it's perilous. Have robots transport us from one padded room to the next, and you kill our joi de vivre.
Yes, because Google (and the authorities letting these cars on the roads) would have *never* thought of the possibility of pedestrians running in front of these cars.
Quick! Get in touch with them and bring this to their attention!
Exactly, I read a recent review of an Audi (I think) that can already automatically brake your car if a pedestrian runs in front of you, at night.
To have a right to do a thing is not at all the same as to be right in doing it
Technically, according to the data they have so far that child would still be safer with the driverless car than a human. It stops for all obstacles, ALL, and THEN maneuvers around them. There are no exceptions, and regardless of anything else, the child is an obstacle.
Time and again it has been demonstrated that human intelligence is biased towards certain kinds of tasks. And it is debatable if driving is one of those
About 40,000 road fatalities a year in the US alone is pretty good evidence that human intelligence is not up to the task of driving.
"Total destruction the only solution" - Bob Marley
How do you test the system for these things? Tens of thousands of hours of real-world driving
=Smidge=
I really hope that Google had some fun with the driverless car on real roads. There's one or two pranks you could potentially pull off. :3
Random Thoughts From A Diseased Mind (Not For Dummies)
...it can broadcast a signal to brake and a whole row of cars can instantly come to a halt...
Insert hack here.
And a human driver uses the superstopbreakpedal thus can avoid the situation better than the computer?
No, but he can see that the car in front of him starts going to the right when emergency stopping, so he can make the decision to go to the left of it. And he can make a choice between running over the child who fell from the overpass or the bag that fell next to it. Choices. Sometimes bad ones, but choices.
(And yes, a really good driver who doesn't panic can stop faster without ABS than with it, so in a sense they have a superstobreakpedal that the robot-driven car doesn't. But that's rather irrelevant here, because it only applies to a very few drivers and cars.)
... for the baby boomer generation to age out of driving. It will now be much easier for all those seniors with upside-down mortgages on houses in developments where you must drive to buy anything - they can stay in their homes.
well even in this case the blind man wasn't really driving and he was accompanied by a police officer. and i wouldn't be surprised if the car had dual controls so the officer could hit the brakes if needed.
all of the special cases you mention could be solved if a human could turn off the autopilot and take over when needed.
i would like it if the autopilot could take over on my normal commute i could have some free time to do something else, or take a nap. that doesn't mean i can't take over if something unusual is going on.
They can also drive safely millimetres (like inches but smaller) apart from each other, massively increasing the capacity of the existing road network.
Stopping distance doesn't change that much. The reaction time becomes smaller, but the braking time stays the same. It's fine for normal use, but when the car in front collides with an oncoming vehicle or something falling off bridge and comes to an abrupt stop then your driverless car still needs almost as long as a human-operated car to come to a safe stop without hitting the vehicle in front.
For one car on its own that is true. The post you responded to is about multiple computer-controlled cars, and for that case, it is absolutely false that the computer needs as much of a braking space as a human driver does. There is no or little advantage for the car in front, but all the other cars in the car-train behind will be able to brake at exactly the same time and achieve exactly the same deceleration as the front car, so the whole car-train will slow down as one. In contrast, human drivers have to notice the car in front of them breaking and then they have to react, and this sequence of notice-react has to happen all the way down the car-train. This is why you should not tailgate, because then you won't have enough time to react and slow down. Computer cars can tailgate other computer cars with no issue, so you have less space on the road taken up by people trying to not tailgate, thus increasing road capacity, which was the point of the post you were responding to.
I can certainly see this working well on highways and other types of roads that are long, uninterrupted, and highly predictable. Especially if there's special lanes for them, like the HOV lanes but smarter. (Insert driver intelligence quips here.)
I don't know how well they would do on the residential streets with all the variables in place. Add on the need for any type of sensors or other fixed equipment and having them off the highway is in a far more distant future.
Would I want them driving in regular traffic? Not so much simply because of the mix of automation and humans. The autonomous cars simply can't cope with the number of variable reactions that humans provide. Not to mention the soon to be developed hobby of messing with the autonomous cars to see how they'll react. I can't believe it wouldn't happen. Then the blame would be on the car, not the idjit who decided to see what would happen if he slammed on his brakes in front of one when traffic is going 50mph.
There's a future in this. Eventually. And every advance is a good one, imo.
That's what they said when factories were invented, or when automation was brought to agriculture.
But as far as I can see, we still have to work, we just do different kind of works.
The interesting part is if we can get past the knee jerk reactions when the autonomous car has hit its first car, has been hit by other car due to confusing behaviour, has killed its first pedestrian, has gone over its first cliff killing the passengers etc because accidents will happen. Accidents will have to be investigated as thoroughly as air plane crashes.
The technology seems pretty wonderful, but like all things made by men, it will at some point fail. Even if the software is designed to handle all possible scenarios and implemented without flaw (good luck), it's still all dependent on physical sensors which are subject to physical failure, sometimes in subtle ways.
Inevitably, if our society permits these to be widely used, we'll see one crash into something (or someone) somewhere. What then?
Who is at fault for the crash? The person behind the wheel? The company that built the hardware? If we write law to the effect of "the person in the driver's seat is always the driver and responsible for the car", that ixnays the blind-driver scenario pretty fast -- too bad for Mr. Mahan! If we choose to designate the manufacturer of the autopilot as the responsible party, expect claims against them Real Fast -- "No officer, it wasn't me that ran over that kid, Google was driving!"
I understand it's not black-and-white -- I favor responsibility distributed to the "driver", but I have personally had my brakes fail dramatically due to improper maintenance by the auto dealer. I would have disclaimed responsibility if I'd crashed after my brakes failed.
However the it shakes out in the courts, it seems sensible to deal with the legalities before the systems become popular -- if they are tested enough, it becomes possible that they'll be widely deployed before the first Real Incident occurs, which will only serve to make it more dramatic. That's drama I could do without.
Will Google foot the insurance bill if I use their gear?
I'm blind, not stupid or incapable of considering the risks and benefits of decisions I make. Just like the guy who sat in this car is presumably just blind and isn't stupid.
Being disabled doesn't mean that I need to be protected from taking risks or that people are being irresponsible by "allowing" me to take those risks.
I might be safer if I didn't live alone, walk through a crowded city every day to work (or have a job at all for that matter!), travel to other places alone, go out for runs to stay fit, or any number of other "potentially dangerous" activities I engage in versus staying in my home and being cared for. I might have been safer if my parents had never sent me to school, or to college, and if they'd never let me move away from home. All of those things would have reduced my risk, but think about the cost.
We all accept risk in our lives. Disabled or non-disabled.
Before you badmouth the people putting blind people in cars, please consider how hard blind people have struggled to get where we are. The level of independence blind people in developed countries have today is better than it has ever been, and there was a time when being blind meant you sat in a house somewhere, and you didn't get a quality education, and you certainly didn't get a real job. There was a time when everyone thought like you, that blind people are nothing but helpless children who need to be protected as though we cannot protect ourselves.
You are not the first person to claim that any number of things are too dangerous, or too complex, for a blind person to do. You are definitely not the first person to be very wrong about that.
I can't think of very many blind people who would turn down a safe and tested opportunity to participate in the development of self-driving cars. This doesn't make them stupid, and it doesn't mean that they're defenseless little children being exploited by a big bad company.
In the future, please let us tell you what we can and cannot do ourselves, thanks.
Ah ha, I was just about to admit defeat and say I couldn't find the page anymore. It is EU not UK. They're apparently well into testing at this point.
http://www.sartre-project.eu/en/Sidor/default.aspx
so in a sense they have a superstobreakpedal that the robot-driven car doesn't
They probably could replace ABS with a sensor that allows the robot to threshold brake as well as any human.
If I have been able to see further than others, it is because I bought a pair of binoculars.
Why would a car collide with anything?
If you cannot stop your car, for any reason, you are driving too fast.
This post contains benzene, nitrosamines, formaldehyde and hydrogen cyanide.
I bet you lived on an time when comercial airline flights left thousands of people in the sea travel business out of work, and i bet you did not make a fuss about it then. Also I strongly believe that Working for work's sake it's not the meaning or purpose of life, when machines take work from us we will find more fulfilling tasks than work to fill all that spare time, you really think humans were meant to type away in cubicles all day, THAT is the purpose in life...really?
That's the real problem. I had hopes for automatic cars being able to fix the issue of having idiots who can't read the 5 road signs telling them their exit was coming up and stopping in the fast lane as they try to cross 6 lanes of freeway traffic to get to their exit in 50 feet.
But then I realized that the robot drivers probably can't read the signs either.
Wonder how these things deal with detours, road construction, etc. Can it recognize the road cones and flashing lights as permanent obstructions its going to have to drive around, or will it wait for the things in the road to get out of its way?
If I have been able to see further than others, it is because I bought a pair of binoculars.
When automated cars form convoys, they are in communication with each other. That means if something causes the first car to brake, the rest of the cars know instantly that the lead car has braked, and that they should brake. Insert "you go left I go right" logic (if needed), and these cars are now better than you at doing it. There is nothing human drivers can do that these cars can't, as their sensor packages are far better than ours, and they can communicate with other cars allowing driving formations humans simply can not handle. The logic they've put into these cars is simply staggering, and that's what's taken so long to develop.
"than a future where all traffic stops and backs up for hours because a branch fell onto the road and the systems raise an alert for "unknown, may not be safe to pass", and don't understand the policeman who waves at you to go around it despite having to cross double yellow lines." easy fix for that have the various redirection things have a transponder/glyph on them so that the code reads
"redirection device detected =TEMP OVERRIDE PROTOCOL" then the system would route around the branch and the Double Yellow safely and send that data up the line so that GMaps can have that road flagged as "blocked/congested".
Any person using FTFY or editing my postings agrees to a US$50.00 charge
Read my post again. If a car in front hits something then it will stop faster than if it brakes. It doesn't matter if the car behind starts braking instantly, it will still hit front car. The car behind will still hit it, and so on. This is how you get a pile-up.
I am TheRaven on Soylent News
Frankly, I have much more faith in a Google computer driving my car then I do the other humans on the road.
Normally I would disagree with such a statement, but after reading some of the comments from the recent poll about where your hands are on the steering wheel, I have to agree.
Some of the comments were downright frightening at how unsafe they were. Holding the wheel at the bottom (six o'clock) with one hand, left hand/right hand crossing the steering to the 10 or 2 position, respectively, driving with knees and bragging about making corners.
I am not for computers driving vehicles, for various reasons, but people such as the ones I mentioned NEED to have the computer do the driving if only to protect everyone else on the road from their malicious behavior.
We will bankrupt ourselves in the vain search for absolute security. -- Dwight D. Eisenhower
My only complaint was that I was not allowed to "drive" this thing. I realize that "driving" it is about as involved as being a passenger in any other car, but man would I want to do it anyways.
If you want to live in a padded room, go ahead. But don't force others to, in the name of safety. Living is dangerous, and that's what makes it worth living.
Then what's wrong with allowing those of us who choose to use driverless cars? Just because you choose the "freedom" to do what *you* want you want to take away my freedom to do what *i* want? fuck you. You must not live in an area like Los Angeles. Every driver here is an idiot and can't be trusted to do what's right (myself included) Driverless cars would most likely make this city much safer place to live.
Uh, what? You're driving along a street at 30 miles per hour. Child runs onto the road 5-10m in front of you. If you hit the brakes instantly, you're still going to hit the child. So you swerve instead. Only there's a cat running onto the road on the other side. You can't possibly avoid hitting one or the other.
I am TheRaven on Soylent News
"The car can easily read these signs (OCR could even be employed) and simply route around it. If a car wanted to make the next left, happens across an orange detour sign, or cones in the way, it will instantly recalculate and just route around it. It's really not that hard to imagine these solutions."
actually OCR is more or less setup with DOT signs (and its a US federal level standard) so combine that with a bit of pattern recognition and we would be set (you would need patrec to account for seeing half of say a stop sign and knowing that it is a stop sign.
Any person using FTFY or editing my postings agrees to a US$50.00 charge
I think that generally, automated vehicles would indeed be safer than your typical human motorist. However, there is one situation that gives me reason to hesitate; carjackings.
A year or so ago, there was a group of thugs in East St. Louis Missouri who, upon the approach of a single vehicle in the 2-3am hours, would run into the road to block the vehicle, and when it slowed, would beat and rob the owner. It's our natural instinct to stop when we see someone in front of us; and even in this case it's arguably the correct response (I suppose it's better to be beaten and robbed than to have the guilt of hitting someone on your conscious). Any autonomous system would be programmed to do the same; making it trivial to predictably stop the car just by approaching it from the front. When this situation happened to my friend, rather than slow down, he downshifted and sped towards them. They jumped out of his way. It was a risky move, but he avoided being robbed and possibly killed.
My (perhaps unwarranted) fear is that when such systems become common, these kinds of crimes will become routine. Even moreso if/when the driver is unable to override the collision detection. And yes, I realize that this is an outlier case, but you know how people tend to give unnecessary weight to outliers that affected them personally.
You go play Russian Roulette with your own life, don't expect others to do the same. If people want automated cars that are inherently safer than the alternative, let them have them.
I've noticed too that most people assume they are better at things then they actually are - not just with driving but in general. There are very few people who can actually multitask, but the majority assume that they are one of those few (even when presented with evidence to the contrary).
Many people swear that talking/texting/browsing/whatever on their phone does not inhibit their driving abilities. The reality is that there is maybe 1 person in 1,000 who is capable of doing so, but that one person is typically smart enough not to do it.
Unfortunately, perception is reality. People's perceptions are that humans are awesome and the best equipped to respond in some hypothetical unknown situation. This clouds people's reception of a computer, which is capable of processing and reacting far faster than a human, that is capable of handling day to day driving.
Are there situations where a computer can't figure out what to do? Certainly, and that's when a human would be needed. But in this case a computer controlled car would pull safely off the road, and alert the driver to take over. Mesh the cars, and eventually all the cars learn how to deal with this aberrant situation.
~shrug~ But I'm an optimist that hates driving. I'm ready to stop.
A person on the other side of the street starts to walk out in traffic, and you see that the oncoming driver will swerve to miss him. You compensate. Will Google's car be programmed to handle this?
more or less standard collision avoidance
A lawn sprinkler at the side of the road is no danger, and its spray can safely be driven through. Is the Google car smart enough to realize this?
would be programmed as "Water Spray NOT BLOCKING"
A trailer is driving very slowly, but signalling with its light that it's safe to pass. Does the Google car make a safe pass, or stay behind?
would be more difficult but could be dealt with as "On Vehicle signal PASS ALLOWED"
Someone has hung rubber ribbons from a bridge to get cars to stop before proceeding. Is the Google car programmed to handle this?
possible how are these ribbons placed??
A piece of road has been washed out, and there are a couple of planks to drive over. Who programmed the Google car to handle this?
should be more part of the base programming
A policeman is directing traffic, and standing with one arm up facing you. What does the Google car do?
A policeman is directing traffic, and standing with one arm up facing away from you. What does the Google car do?
pattern recognition should solve that part (first case i think is STOP second is GO)
Any person using FTFY or editing my postings agrees to a US$50.00 charge
And then there's fecking ignorance of the scale of work being done.
I went to a lecture about autonomous cars in grad school -- twenty years ago. This is not a field that Google suddenly invented and started working on, they are building on decades of robotics research. It's not just Google's measly 200,000 miles.
Tom Swiss | the infamous tms | my blog
You cannot wash away blood with blood
During a Google talk, they mentioned that the front radar on the autonomous car can detect objects in front of an 18-wheeler tractor trailer from behind said truck.
Google's autonomous cars are capable of things humans would never be capable of. Its time to embrace the future, just as we did when we said goodbye to horse-drawn carriages.
The Motorcycle Safety Foundation regularly tests professional racers using threshold braking against normal riders using ABS, and the ABS has been the winner in stopping distance for several years now.
In order for Computers to be driving us on roads, we will have to remove all non-computer drivers, otherwise Humans and Computers will conflict with each other.
-- By all means let's be open-minded, but not so open-minded that our brains drop out.
"...Technology won't have a blind spot."
No. No, it will not. Anywhere.
Speaking of which, does it bother anyone beside me that Google wants to put multiple cameras on cars around the world? Can anyone really expect these cameras will not be put to other uses? Is data obtained by those cameras stored in any way? Who owns that data, and is it admissible in a courtroom?
I am beginning to hate technology. For every advancement in technology, there seems to be a corresponding loss in privacy rights and liberties.
What if there's a child running out from one side of the road, a war veteran in an out of control wheelchair sliding on to the other side of the road, a cat jumping out of a tree into the middle of the road, an asteroid striking right behind you, and a tsunami coming from in front of you? What then, huh?
Is 1563649 a prime number?
a wheelchair isn't a two ton death machine controlled by computers hurled through the universe.
Still safer than a two ton death machine controlled by a human.
The sooner we can get those things on the road the sooner we can get rid of a lot of unnecessary deaths. (Probably not all, but one can always hope.)
I'd bet a good portion of human drivers in that situation would hit the child.
Why wouldn't they? They're perfect?
So basically you are arguing for killing 6.5 million people a year, simply because driver-less cars will not reduce that number to zero but "only" to under a million.
I really have to question your morals when you actively wish death on six and a half million people each year for no good reason what so ever. In fact your desire to see people hurt maimed and killed pretty much negates your arguments, since clearly you do not have any human beings safety or life in mind.
Human level AI is so much more incapable harmful and damaging compared to current computer AI is so obvious that one can only assume you actually are desiring these people die while mangled in crushed and burning metal, since you argue against anything that would lower that number even in the slightest, let alone the huge amount this actually is.
You sicken me.
They may hit the child, but they'd place trying not to higher than trying not to hit the cat. If the car can't distinguish between car and cat, it wouldn't swerve and so it would plough straight into the child and at the inquest it would be shown that it didn't even attempt not to. This would be a major PR problem for the manufacturer.
I am TheRaven on Soylent News
Uh, what? You're driving along a street at 30 miles per hour. Child runs onto the road 5-10m in front of you. If you hit the brakes instantly, you're still going to hit the child. So you swerve instead. Only there's a cat running onto the road on the other side. You can't possibly avoid hitting one or the other.
The car can see the difference between a human and a cat. I don't know if it can see the difference between a human and a monkey dressed as a human.
I'm much less worried about a driverless car hitting a pedestrian than I am the average driver hitting one.
Even if the car's automation has Microsoft Windows Automotive 10 running on it?
A lot of the resistance to the idea of automated vehicles is 'who is responsible if there is an accident'. If one ever does get in an incident, especially the first couple of times, I expect the payouts to be higher than for an equivalent human-caused incident. I think it is easy to measure the safety performance of driverless cars, and I expect that it is better than regular drivers, and will improve. That means that insurance for driverless cars should be cheaper than for human drivers. So Google should offer to insure all their driverless vehicles, and because those vehicles are safer, they will come out ahead.
This is such an incredibly rare situation that I think we can comfortably let the designers worry about it later, and enjoy the dramatic decrease in loss of life that the driverless cars bring us right now.
I completely agree with this, but I wonder if society will in general. It's all fine and dandy to say that human drivers result in X fatalities per year, and robot drivers result in half that. You can have statistics and charts and 8x10 color glossy photos with circles and arrows drawn on them and a paragraph on the back of each explaining what it shows to back you up. That's not going to stop someone from bringing a now-quadriplegic kid who was crippled by a johnny-cab into court and suing the manufacturer into oblivion. The jury won't care that statistically the robot drivers are safer. They'll just care that in this instance the robot car caused grievous bodily injury. Never underestimate the power of an injured child in a courtroom.
Chelloveck
I give up on debugging. From now on, SIGSEGV is a feature.
You take an insane situation LOTS of humans would screw up, but you have no idea if the car would screw up, then say the humans would do it perfectly while the automated car would always screw it up.
You're insane if you think you're making a sincere argument.
We know how stupid humans are, how could we create anything smarter than ourselves?
You've never met a smart kid with stupid parents?
What about kids throwing stones off the top of a bridge onto the passing cars (common problem in the UK - someone died just the other month from this)? Is the car looking UP too and determining their intent?
By your own example, humans already fail at this. Hardly the end of the world if the car doesn't do it either.
Counterpoint: Formula 1 banned ABS brakes after initially allowing them.
But that's neither here nor there - for a typical driver, ABS is going to be a benefit. There are few Nigel Mansells on the road.
The AI quickly uploads itself to the Googleplex so it can be installed in a new unit.
-=This sig has nothing to do with my comment. Move along now=-
Thanks, hope somebody will mod you up
I do not believe in karma. "Funny"=-6. Do good and forbid evil. Yours, Oft-Offtopic Flamebaiting Troll.
Then what's wrong with allowing those of us who choose to use driverless cars? Just because you choose the "freedom" to do what *you* want you want to take away my freedom to do what *i* want? fuck you.
As long as your driverless car doesn't impede my driving, I see no problems. The problem is that it will.
Counterpoint: So you would have no problems with drivers doing 10 mph on the freeway then? After all, that speed is much safer, and by your reasoning, you don't want to take away their freedom to do what *they* want?
I dunno about you, but I would feel safer running in front of a robot controlled car than a human controlled car. Of course, the first robot car accident will bring happy lawsuit time.
This is going to be the biggest hurdle to self-driving cars anytime in the near future. It really doesn't matter whether it does better than people will. Most people will get far more upset about the self-driving car hitting their kid than a person hitting their kid. The self-driving car will probably have to do 100x to 1000x times better before most of the population will really accept it.
The reaction time will become much smaller. We used to call it PIEV (Perception, Intellection, Emotion, Volition), but recently has been referred to as PRT (Perception-Reaction Time). Both terms are explained in this paper. In a nutshell, though, this process can take 1.5-2.5 seconds depending on the person, circumstances, et c. That doesn't count the stopping distance, which varies widely by person. Total PRT and Stopping distance from 65 mph will vary from 350-450 feet, depending on circumstances. As PRT approaches zero, the total distance traveled approaches the stopping distance, probably about 200 ft, depending on the vehicle.
Also, the post you were responding to assumes that the cars can communicate
Not if it means that the average speed of transportation goes way down.
Current projections expect that overall capacity with full automation would double over current capacity, assuming that allowed speeds don't change.
I want to see the CVs of these engineers first. On how many high-integrity systems have they worked so far? I'd also need to know which programming languages and development tools they have used, see the source code, and would like to know which formal software and hardware verification methods were used to verify the code.
Um, to put it simply: No.
You do not warrant that level of information (in fairness, neither do I). Do you question the certifications of the people who installed the software on your current car? Do you not trust the person who installed your airbags, or their associated sensors, expecting them to deploy at any moment as you drive down the road? Do you disregard the "check engine" light, assuming that without knowing the sum history of the people who designed that light, it's probably inaccurate?
Do you question airline pilots credentials before boarding a plane? Do you need to see the CV of every bartender who has served you a beverage, every waiter or waitress when you go out to eat? Do you personally get to inspect the kitchen of every restaurant you visit, and are you given a complete life history of each chef and the pedigree of the cow which provided the beef for your meal?
Any one of those things could easily kill you or cause grievous bodily harm, yet I suspect you allow those people to do their jobs and extend them common courtesy of not assuming incompetence. Why don't you extend the same courtesy to the engineers working on this project. For reasons which should be apparent, google will not be tossing some slap-dash machinery out the door. It will be thoroughly tested with hundreds of thousands of road hours before you ever have the opportunity of sitting in that seat.
And should the worst come to pass, and something does cause you grievous bodily harm, will you ask for the work history of the trauma surgeon who attempts to reassemble your broken body?
This signature is false.
You know what I do not like about it? It's that it seems oriented to convienience of the driver. If you ask me, drivers are already too comfortable with power-everything, so they start driving sloppily.
I would rather see those trains geared to high speed traffic with special lanes etc.
Right now it looks like those trains will be riding at the speed of the bus. We already having buses riding HOV lanes at the speed lower than everything else.
I do not believe in karma. "Funny"=-6. Do good and forbid evil. Yours, Oft-Offtopic Flamebaiting Troll.
I want to know how did he know what was on the Taco Bell Drive-in. Don't tell me the car read him what was on the order board!
Paul: Father... father, the sleeper has awakened! - Dune
Automated cars are also unlikely to rip along at 80-90 kph in a 50 zone like the psycho-cabbie who nearly ran me down on my wake-up walk half an hour ago, too.
So, let me get this straight... You're half asleep while walking on a 50 kph road?
And the driver is the asshole?
Doesn't that depend on the girth of the rider? I've seen some wheelchairs and some of those "scooters for the obese" that needing reinforcements at their welds due to folks that would take up two full seats anywhere else. I'm still traumatized by that one time I was flying (before 911 and the irradiating scanners and pedophiles groping everyone at the airport) when I was forced to share half my seat with a person that had to arrange their rolls to fit the armrest between them. I didn't see the armrest for the entirety of the flight. It was entombed entirely.
Some folks don't need more automation in their lives, they need less.
That said, I'm all for the cages being automated, so that the monkeys inside them can put on their make up, chat on their cell phones, text their friends, and drink to their hearts content, without impinging on my right to ride on the road as well.
I want to see the CVs of these engineers first. On how many high-integrity systems have they worked so far? I know plenty of people with an AI background, and trust me, I don't want these to program my car. I'd also need to know which programming languages and development tools they have used, see the source code, and would like to know which formal software and hardware verification methods were used to verify the code.
I'm fortunate enough to know some of the people on the team. They come from robotics backgrounds and are very experienced in highly reliable systems. For example, a big chunk of the Google team was hired out of the Carnegie Mellon Tartan Racing team. That team tested their systems so much that they wore out hardware. The Carnegie Mellon team also originates from the Field Robotics Center which sends robots to Antarctica, volcanoes, sinkhole lakes, the mountain deserts of Chile, and abandoned mines. These are hardened systems developed using hard core systems engineering and serious software engineering methods. Trust me, this is not a group of ivory tower AI people ignoring version control and using latest beta release of LISP. One of the technical leads, Chris Urmson gave a talk last year at Carnegie Mellon. It was apparent that they have continued these practices in Google. The incredible volume of field testing is the outward evidence.
As long as your driverless car doesn't impede my driving, I see no problems. The problem is that it will.
People already impede your ability to drive so that argument doesn't hold water. I wasn't arguing safety at all. In Los Angeles we already drive 10 mph on the freeway but laws are already on the books about impeding the flow of traffic and so that countpoint fails as well. I am not for safety - I'm for self-driving cars.
In Soviet Google car test drives you
Technology doesn't get distracted by cell phones, GPS systems, or the radio.
Some technology does.
The problem with this is that obstacles don't just appear out of thin air; the majority of the time that a car rams into an obstacle without breaking, it's due to human error (following way too close to react, being too distracted to react). This is much less likely to happen with a system that is programmed not to follow too closely and doesn't become distracted, plus is sensing its surroundings many more times/second than a person would.
No one is saying that this will prevent all accidents, or that there aren't other issues that can crop up (sick idiots figuring out how to feed a system false information, malfunctions of the sensors, etc.). But a properly designed automated system starts off being better functioning that what we have now, which is people of varying intelligence and ability, many of whom feel an entitlement to drive and have no sense of how to work with the other traffic around them so everyone can get to where they're going.
I agree - computers can play chess really really well. Better than 100% of humans. But that is a mathematical game that needs no messy inputs, and has what 30 possible moves - which need to be done once per minute.
Its kind of obvious why google is doing this - not Ford or Honda - because the people at Ford know how to build a car, and the complexities of testing. A few hundred thousand miles is nothing for Ford to test a single new brake pad design - which is likely some super minor change. They test stuff like that millions of miles before bolting on in the assembly line.
The lawsuits would be spectacular! Google killed that pedestrian, not the driver....
Hilarious choice of testing facility - southern california... no weather.
On a related note, I always thought traffic congestion must waste an enormous amount of fuel. Years ago I came across a stat related to this done by a Texas university, which stated that the daily average was something like 250k barrels of oil per day - which may seem like a lot, but bear in mind the US plows through ca. 9 million barrels of gasoline per day.
My original link is now a 404 but I've found the data source they used: Congestion Data for Your City — Urban Mobility Information. Go to the data for "All 439 Urban Areas," and down to the table for sums. This states that due to traffic congestion 1,943,330,000 gallons were wasted for 2010, thus 5,324,205 per day; there are 42 gallons in a barrel, thus 126,767 barrels/day. I'm not sure if this is the correct way to calculate this, as refineries actually crank out more like 19 barrels of gasoline per full barrel of crude; so maybe 280,221 b/d? That must be how the original study came to its conclusion. Either way it's not much of a dent in the 8.71 million barrels of oil we wolfed down in the form of gasoline last week: Petroleum and Other Liquids - Data - U.S. Energy Information Administration (EIA)
Yeah, but then you have to clean off all 50 cameras before you drive your car in the winter. Currently, things are pretty simple. Clean off hte windowshield (ice melted by car heater), melt the ice off the back window, scrape the ice off the windows enough so you can see. But now you have all these little cameras you have to make sure are clean. What happens when ice freezes on the lense? Is it easy to clean the ice off the lense without damaging it?
Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
Right, because human drivers have never hit a child that ran out in front of them. It's a shame really, according to some quick searches about 300 kids get hit by cars every year. Most because humans can't see in all direction and angles at once (most are in driveways and many by the parent). If we switched to automated cars tomorrow and 100 kids got hit next year, people like you would blame automated driving as the problem. Even if it was 1/3 as many incidents.
It will never happen in first-world countries, for one simple reason: litigation. Even if automobile fatalities were reduced by 50%, all it takes is one dumb kid running out into the street, causing an accident that neither human nor robot could avoid. Then the manufacturer and every other company in the supply chain will be sued for millions upon millions. Lawyers will have a field day. Insurance rates will skyrocket. Companies will refuse to shoulder the liabilities, so they won't even make them--it's just too financially risky. Even if they are statistically safer, that won't matter in court when an accident does happen (and they will, because nothing and no one is perfect). A human driver is responsible for his own driving, but the manufacturer will be responsible for its robotic driving, and no car manufacturer with lawyers in their "right" minds would allow such a thing to proceed.
Maybe it could work in other parts of the world, places with less-developed legal systems....
"Those who consume the bulk of goods are those who make them. We must never forget this secret of our prosperity."
If it was truly smart it would stay in the garage and never come out.
[insert clairvoyant elevators joke]
Why does everyone on Slashdot keep on bring up driverless car convoys? I can't be the only one that think they sound like a really bad idea. Seriously, I want my car to trust its sensors, not what the car in front of it is saying. I can't rely on human drivers to use turn signals correctly. You want me to trust that the car in front of me is both programmed correctly and not malicious?
well soon in GOP USA people in lockup get healthcare care with fully covered vs working 39.5 hours a week with no plan on the out side.
autopilot software goes under FAA review and they just don't let any piece of code go into the code base.
even long term there will be a switch over period with auto cars and non auto cars on the road at the same time and likely there will be manual drive only areas and or areas where you can't wall off pedestrians. (maybe you can but very few areas have las vegas pedestrian overpasses)
They can find jobs maintaining the cars!
Plenty of people don't find a purpose for living now. How is this different?
In the immediate future, the states that do allow "the google car" requires the driver to be fully licensed, attentive, sitting in the driver's seat, and not-drunk or visually handicap. I don't see this changing anytime in the near future as this technology has a long way to go to be "perfected" but on the good side of all of this, this car can already avoid unforeseen obstacles such as humans jumping in front of the car. However, we all know that stupid people will try and "test" these machines and get hit in the process. For liability reasons, cameras/sensors everywhere will be needed in and out of the car.
Yep everyone else needs self driving cars, but not you. Your special.
I am beginning to hate technology. For every advancement in technology, there seems to be a corresponding loss in privacy rights and liberties.
This has nothing to do with technology itself. It has to do with the people who use the technology. For example, your internet history does not need to be logged in order for your ISP to provide you service. GPS devices do not need to send your currently location to some server somewhere in order for you device to know where it is. Social sites do not need to data mine your information in order to function. The loss of privacy is due to the fact that most of society does not care enough to do anything when companies and individuals take advantage.
Better yet. Take his free online class. http://www.udacity.com/
I would rather have an accident caused by human mistake every now and then than a future where all traffic stops and backs up for hours because a branch fell onto the road and the systems raise an alert for "unknown, may not be safe to pass", and don't understand the policeman who waves at you to go around it despite having to cross double yellow lines.
This is why you have overrides.
I'm surprised there's no discussion of some of the most interesting points. Namely, if automatic cars are commercial and commonplace, then who's really in charge of them? Considering the potential danger of any errors in any changes to the control software, I'd have to expect it would be protected from any changes every way you could imagine (TMP, signed bootloaders/OS, all of that). Or else some inept hacker/terrorist/total nutcase will screw it up and cause some huge wreck.
So Google or Ford or whoever will probably be the ones in control of the software, and therefore the ones really in control of what the car does. And large companies tend to do whatever the Government asks them to, because they have a lot to lose. So, how long until it's set up so that if the police want you for any reason, good or bad, any car you get in will take you directly to them, refusing to do anything else? Probably all of the cars will also keep records of everywhere they go, and maybe everyone who rides in them too. And all those records will be in the hands of Google or whoever, available to any police agency with, or maybe without, a court order. Maybe available in bulk, for whatever bulk analysis anybody can dream up. Want a list of everyone who drove by that street where they sell drugs? Everyone who drove by a crime scene, just to see if there are any witnesses or accomplices? Here you go. What else can you dream up that someone might do with all of that data?
I do see the tremendous number of lives that could be saved and injuries prevented when human error is taken out of driving. I also think it's worth considering what else we might be giving up by doing it.
I don't reply to ACs
Replacing roads with rails, ....
Actually, I kind of wish they had rail systems along the interstates that you could attach to and detach from. Semi-trucks and long distance drivers could connect to the rail and not have to get off until their exists several hundred miles away. While on the rail, the speed/brake would be controlled by computers in order to avoid collisions.
All the machine has to do is make fewer mistakes than the human. When it gets there and you still take issue, the problem is you.
Actually, the machine only needs to make, at most, an equal number of mistakes to make sense from the safety perspective. Of course, when I say "equal number", I mean the number of mistakes weighted by the impact of those mistakes.
Why would we force people to use driverless cars? Even if we thought everyone should use them, we wouldn't need to force anyone. Many people will see the utility of not having to drive and want to such a car. Assuming it is affordable, people will buy the cars. Some of those people will be bad drivers which will make the roads safer since they are no longer driving. I could see forcing people with so many points on their driver's license to use driverless cars.
A human would typically prioritise not hitting the child.
Traumatising a kid for life as you run over it's favourite pet, vs killing the annoying 2 legged thing who keeps picking up and touching me while I'm just trying to rule the house from my comfy nap point.
I think the cat may disagree with you :)
From one of the videos from the Udacity class, apparently they still not have solved the snow problem. They talk about possibly using trees, buildings, and such as reference points instead of the lines on the road.
Tons of lame excuses in this thread for not accepting the start of a great technology. I have been saying for years that I wish my car could hook up to railroad tracks and autopilot me along them. This system would be so much better. Most of the complaints are from not RTFA. I know, must be new here. The rest of the complaints most humans cannot even deal with. The amount of energy that it would take to be as alert as these sensors 100% of the time would probably kill a normal human or at the very least, the humans senses would quickly degrade. I think a great start to this would be patches of straight, uncluttered, highway that is designed to handle this type of vehicle. Go from there! Sign me up!
has gene hackman rigged the car to explode should it slow or stop?
Don't forget about insurance costs. If these things really are statistically that much safer, insurance companies will be all over that in no time at all, and will lower rates for their drivers (or, perhaps, rather raise rates for drivers of "manual" cars). That can be a pretty strong incentive.
Actually, no - it uses voice recognition to detect whether it's in the South or Midwest based on the prevailing dialect it hears, and then enables the feature.
If/when it gets to that, we'll need to change the way our economy functions. When we only need 10% of the population to work (and in work actually requiring human intellect at that), the model of "must work to earn a living" will not be functioning anymore. At which point we'll probably move on to something like guaranteed minimum income, using up wealth produced by all that automation. I don't see anything wrong with that - it's not like you need to work for a living to find a purpose to live.
There is quite a significant difference in distance between a millimetre and an inch.
More like rare weather. In Southern California, the moment there is a slight drizzle the drivers don't know what to do, and lose whatever ability to drive they might have had when it was dry.
They can also drive safely millimetres (like inches but smaller) apart from each other, massively increasing the capacity of the existing road network.
You don't know that for certain. AFAIK they have never put two Google cars together on the same road.
Which makes sense, when you think about how laser range finding works. If you put a second car with an identical setup next to it, you would surely get crosstalk between the two rigs.
sub f{($f)=@_;print"$f(q{$f});";}f(q{sub f{($f)=@_;print"$f(q{$f});";}f});
It's not a matter of the situations one can easily anticipate, such as pedestrians in the road, but rather the ones that havn't been anticipated. You may not have noticed, but human behavior is rather varied and can't simply be enumerated as a checklist.
For example, if a human driver sees a ball rolling out into the street, you know to slow down immediately since there's a good chance there's a child about to come running out after it, but the computer -if it hadn't been pre-programmed for that scenario (or had human level AI) - wouldn't react until it saw the child, when it may be too late to brake.
So you program that one in... but forget to program the fact that an oncoming driver may swerve to avoid an animal, etc, etc. In general it's the cases where you need to predict human behavior (not react to current conditions) that are going to be a safety nightmare.
The trouble is the real world is messy and unpredictable and you can't simply enumerate and pre-program optimal responses for all situations - you ultimately need human-level AI to fallback on if you're not going to fail in cases where a human would not.
Nope. The problem isn't reaction time, but rather predicting human behavior. The automomous car would likely only break if it saw the child running out into the road, by which time it may already be too late to brake.
The human has the edge because they can also predict when a dangerous situation is going to happen and react before it does. An autonomous car whose intelligence amounts to an expert system rather than a advanced general intelligenve (AGI - yet to be achieved) is always going to have situations where it fails because the situation didn't fall into it's playbook.
In Mountain View, California, Google car drives you!
Also insert beneficial "hack" uses:
Clear the road for Police Chase
Clear the road for Fire Truck
Clear the road for ambulance
Stop car being used by gas station robber too stupid to realize the car can be controlled remotely.
Auto Park, parking lots and street parking designed to advise the car of the nearest empty spot to the destination.
Simple wireless traffic diversion signals for Police, Fire & Rescue Crews, Construction works, parades, protests, whatever. Need to mark a detour? Little beacon they place on the road makes sure all the cars know the road ahead is a no go zone.
XML - A clever joke would be here if
Your assuming that we cant build a self cleaning / de-icing system.
We can.
Ships have centrifugal window ports (the little round windows in another window on the front of some boats/ships) to repel water with salt & grit in it without scratching the surface like a window wiper.
We have developed a LOT of ways to de-ice systems.
Dust can be cleared by electrostatics or with good old fashioned water.
A full self cleaning all weather (rain, hail, ice, snow, mud, dust, etc) system may be complex and little more expensive than just flat glass/plastic but is capable of being built with off the shelf technology today and further innovation will no doubt develop simpler & cheaper ways of integrating these mechanisms to make them better in the long run.
XML - A clever joke would be here if
Some of the comments were downright frightening at how unsafe they were. Holding the wheel at the bottom (six o'clock) with one hand, left hand/right hand crossing the steering to the 10 or 2 position, respectively, driving with knees and bragging about making corners.
What's frightening is that you don't seem to realize that what's important is whether the driver is in control of the wheel and can take corrective actions in either direction. That's what matters.
The standard 10-2 position is not always the safest - when doing gear changes, for example, it adds a natural inclination to swerve towards oncoming traffic - the most dangerous direction. Due to the weight of your arms, it's always easier to pull the steering wheel down than to push it up.
With frequent gear changes, a position where the elbow rests (or, for a small steering wheel, a top position) is safer.
The 10-2 position was invented back at a time when steering wheels were huge and at a 45 degree angle, and cars were without steering assist and massively understeered. The 10-2 position allowed you to rest your elbows on the wheel, and pull it the large arcs needed for the turns.
These days, wheels are much smaller (we have steering assist), and much more tilted towards the driver, we go faster, and a lower position is almost always preferable.
3-6 is the new standard, but many steering columns won't allow it, in which case 4-8 is still better.
Try to think a bit before your next mouth frothing outrage. Examine why someone says something, and not just condemn it because it goes against what you see as conventional wisdom. What "everybody" knows to be true often isn't.
(correction: 3-9 is the new standard, obviously, not 3-6. I turned the wheel too far and the 9 tilted over)
Blah, blah, blah. I've been driving a stick since day one and if someone can't compensate for the extremely slight drift that might happen when shifting gears, they have an issue.
Second, reading comprehension is your friend. I said people were driving with their right hand at 10 or left hand at 2. In other words, they are crossing their arm in front of their body which limits movement and control. Further, this is exactly what the proposed new driving position is trying to prevent: people having their hands/arms in the way if the airbag deploys. This also applied to people ghetto driving: arm straight out, hand at the noon position.
Further, I watch people who drive like this and they are generally weaving from side to side going down a straight road because of the inherent push/pull of that driving position which endangers those around them. When they make turns, they go through all kinds of leaning and adjustments because they have no control with their arms across their bodies.
So when you're done frothing at the mouth, you can reread what I said and stop trying to justify people who believe they're the best drivers in the world and common sense doesn't apply to them. There's a reason professional drivers don't drive the way these people claim they're so good at. It's because there is little control driving with your arms across your body or having one hand at the top or bottom of the wheel.
We will bankrupt ourselves in the vain search for absolute security. -- Dwight D. Eisenhower
Yes, I am special. I can write correct English.
And yes, I don't need a self driving car. I'm not a lazy, incompetent git who thinks driving like they're from the ghetto while holding their cell phone in one hand while trying to text isn't endangering those around them.
I don't mind driving. It's the idiots around me that make it annoying. If people would use one iota of common sense, the vast majority of issues would be resolved.
Instead, we're trying to find a technical solution to a human problem when if the humans would use their supposed intelligence, this wouldn't be an issue.
We will bankrupt ourselves in the vain search for absolute security. -- Dwight D. Eisenhower
Agreed, many good uses. It would put all the valets out of work and solve a lot of downtown parking problems since it could drop you off, go park in whatever ramp has space, then come pick you up again when you use an app to call for it. Hmm, now that I think of it, it could make car sharing much more practical. I still worry about the bad uses though. Like any power, government control of the transportation system will be abused. If you happen to have a name similar to someone who talked to someone who works with a terrorist suspect you may find yourself being driven in for questioning when you thought you were on your way to work.
I believe Google presently relies on hyperaccurate maps (previously collected laser scans) of the route. It seems unlikely that they would go through the trouble of collecting this data if the route was inappropriate. In the future, it seems likely that they'll want the cars to adapt to roads that have never been traveled before, but how often is it that you happen upon a road that lacks Street View coverage (which seems like a great "vehicle" for collecting the laser scans)? Even if self-driving cars only work on (a) highways, (b) major arteries, and possibly (c) routes that you've manually driven before, that's still a huge net win, even if it means I have to take control whenever we leave the self-driving routes.
Many years ago, while I was assigned to the Air Force Test Wing, we had a cargo jet outfitted with a computer flight system that took off, flew and landed by itself. It was very accurate, and very safe. The biggest issue was the pilots assigned to sit in the cockpit in case of a computer issue (who had all kinds of monitoring devices attached to them) had high blood pressure and heart rates during the flights.
How does the car know to stop at the drive-thru speaker? How does it then know when to go again and pull up to the drive-thru pay window? And same again for the pick-up window? And later in the video, how does it know that it's time to choose a parking space? So clearly there is some human control, none of which could be done by someone who is actually fully blind? So in this case, was the 95%-blind person doing something (gas/brake, voice commands), or was a sighted person controlling it in some way? Basically, how does a driver (sighted or not) communicate decisions to the car that are beyond the scope of "driving"?