Slashdot Mirror
Another Step Towards the Driverless Car
jtogel writes "At Essex, we have for some time been working on automatically learning how to race cars in simulation. It turns out that a combination of evolutionary algorithms and neural networks can learn how to beat all humans in racing games, and also come up with some quite interesting, novel behaviours, which might one day make their way into commercial racing games. While this is simulation, the race is now on for the real thing — we are setting up a competition for AI developers, where the goal is to win a race between model cars on real tracks. As the cars will be around half a meter long, the cost of participating will be a fraction of that for the famous DARPA Grand Challenge, whereas the challenges will be similar in terms of computer vision and AI."
At least now they won't cause accidents.
Seriously, this is a technology whose time has come. Persuading elderly drivers to give up their cars is difficult, and the baby boom generation is putting a lot of people in that situation in the next decade or two.
equals bullshit. And I ain't no troll.
I for one welcome our new automated race-car overlords.
I wonder if a driving android can be constructed.
*RUNS*
-uso.
What you hear in the ear, preach from the rooftop Matthew 10.27b
....car drives you!
Somehow I doubt that using my actions during a racing game where I don't actually die and can push beyond reasonable limits on actions is a good source for data for driving simulations.
... why not use GTA: San Andreas for the data?
I know
-- Tigger warning: This post may contain tiggers! --
Persuading elderly drivers to give up their cars is difficult, and the baby boom generation is putting a lot of people in that situation in the next decade or two.
I find backing over them works fairly well.
-- Tigger warning: This post may contain tiggers! --
I would love to have a driverless car: let me get some work or reading done while lounging in the back seat (safer) of my car while it is driving me through the daily rush hour. Because I can get work done, I can either drive off later or am in less of a rush to get where I'm going. No more tedious trips of hours upon hours of driving.
/. I think) said humans have trouble paying consistent attention to anything for an extended period of time without having our minds wander:
n dex.php?action=fullnews&id=80157
Lower insurance premiums - and if the car has an fender bender, I can point to the manufacturer and hopefully won't be branded as an unsafe driver for life if I didn't do the driving. Safer roads for all. A recent study (posted on
http://www.localnewswatch.com/skyvalley/stories/i
We'd be more safe.
No traffic tickets - the AI can go closer to the speed limit than I have the patience to (now if they didn't consistently set the speed limit too low in a ton of places just to be asshats and be able to write tickets when they need the money....)
Seriously, this isn't just for the elderly. Driving ceased to be fun for me long ago. If I had to do it only once a week on a nice stretch of fast highway, I might feel differently....
What's wrong? Get stuck behind a driver who understands the concept of a speed limit?
Seriously, you can complain about elderly drivers once there's some sort of infrastructure in place to ensure that they don't *need* to drive themselves about.
I recall the Forza 2 team commenting just recently that their AI drivers were being powered by neural networks which they'd managed to train to the point that they were actually using some extremely cool braking and steering behaviors that they had never been taught. This seems quite similar.
http://rars.sourceforge.net/
Anyone have first-hand experience of how that is coming along? Can you program these with neural nets also, or still "just" hand-coded? (My own experience is nigh on a decade old.)
"Good news, everyone!"
No no no! I don't want a driverless car, I want a car with native linux drivers!
Who doesn't want a scandinavian driver?
Thats great and all, but will the AI-enabled cars also come with an afro wig and a black leather jacket? Oh, and they should probably be bulletproof.
The theory of relativity doesn't work right in Arkansas.
In Soviet Russia, car drives YOU.
The links go to an AI presentation of virtual cars and a news release saying there will be a competition without any details about the competition. Did I miss something? Is the only news that they have developed neural nets to drive a virtual car?
The competition sounds like a manageable project for academics (versus the DARPA event).
Is the competition still in the vapor-ware or maybe-someday stage?
Anyone have a link (perhaps IEEE) that has details?
Being legally classed as being blind, I would love to have a driverless car. I would live the independance it would bring.
The courses he uses remind me of courses from Super Off Road.
Will these complex neural networks be running on Linux? Or can we expect Windows sludge?
This might sound like the windows based car....
"Sometimes the outcome of the race is that both cars find themselves stuck against some wall."
A few months ago I was thinking about doing something like this, but in heavy traffic situations. What if the course had way more cars than should ever actually fit, and the cars independently tried to minimize their travel time around the course. I wonder if the computer could get a better overall throughput than people seem to do on the crowded highways of, say, Seattle.
Make it either Opt-In or by Court Order
Now this, to me, is a very important distiction. What if for this to work well, all the cars have to be computer controlled? What if computer control is then mandated? This is a whole new exciting level of "nanny government". Sure this might be safer in that there would be fewer auto accidents, but do you really want all transportation to be centrally controlled? Sure each car might be autonomous at first, but emergency workers need the ability to remotely turn on off, right? It's for everyone's safety.
We are all just people.
If anyone actually is interested in reading the papers discussing the experiments we did (many more than you see in the videos!), most of them are available on my website.
Some of them are of course better than others. I can recommend this one, about evolving general and specific driving skills, this one about co-evolution, this one about different learning techniques, and this one about modelling human driving and evolving tracks. There are several new ones, including one on physical cars, which are not on the website yet - mail me if you want a preprint!
All this assuming that anyone actually reads academic papers... sometimes it seems that not even the guy who writes the paper actually reads it. (Not true in my case, of course!)
Sitting in a car with my missus driving is much the same as being in a driverless car:
Biggest difference is that the thing is more likely to know the way to someplace.
Do it yourself, because no one else will do it yourself. [beta blockade 10-17 Feb]
If we start using driverless cars, the Blue Screen of Death will suddenly have a whole new meaning!
if this comment isn't informative, i don't know what is
The problem with driverless cars isn't the technology but insurance.
Many manufacturers have been dissuaded from pursuing the technology and installing in their vehicles because in the case of any accident the corporation would be liable. Obviously the 'driver' wouldn't be at fault because they wouldn't be driving.
No large corporation is going to put itself in line to pay out on every bump, scrape and minor slaying caused when their killer robo-cars Attack!
It is bad enough seeing a steady stream of cars and SUVs with only one person in them streaming out of the downtown at rush hour. Now we're going to have cars out there that aren't even taking anyone anywhere.
The environmentalists will not be happy with this development!
on a real track, and these algorithms wouldn't stand a chance. Racers know subtle moves and blatent moves that these systems will never be able to learn. Add the fact that real cheating and bending the rules has to occur under the nose of race officials, and that team cars run by algorithms would be banned from any racing venue for being dangerous morons within one or two races, and this would disappear faster than most vaporware.
Implying that these cars could "drive themselves" in any meaningful or safe manner is idiotic. I would like to see what would happen if they put in place some rules based on sane driving.
tom
autocrosser and road racer
I hate sigs, and refuse to have one.
there are some things that i really would not want to have to write into the code for a driver less car, but that would need to be addressed. What happens if you are driving along a road and someone jumps out in front of you while you are passing a car going the opposite direction? Does the car swerve to the right to attempt to miss the person, and risk running itself into an object that could kill it's driver? Does it swerve to the left and hit the other car? Does it drive into the person? What about an accident that happens on a free way at rush hour at high speeds like a blown tire, or broken drive train? The chances are small, but things like that will have to be written in. I would love to be able to get in my car, tell it a destination and go, especially if i could go out to a bar, put back a few too many, and not have to worry about how i am getting home. I just don't see it any time soon, sadly.
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0
The article mentioned the cars getting stuck in the corner. This is a problem with the NN architecture used. Backing up requires a very different sort of behavior than normal driving, so you require a way for your architecture to keep track of a particular state, and change behavior as you progress through that state (you need to back up for a set amount of time, then go forward and change direction). This problem is made clearer when imagining the absurdity of trying to get a simple NN like this one to learn how to parallel park. The solution is to add a node or two which maintain their values over time (i.e. they are not directly fed into by sensors) in order to set the car into different states. So, after evolving, one of the nodes will come to represent "back up" state. When the car gets stuck in a corner, the backup node will become activated, and will stay activated for long enough for the car to actually back up! I shouldn't have to mention that this is probably a difficult thing for the car to learn on its own, but without these extra memory nodes it is impossible.
TORCS is a more advanced racing simulation than Rars. Its held robot-programming contests for the last 3 years, with another about to start soon.
There have been several robots that use various learning techniques, though none to my knowledge have been full-blown AI/neural net solutions. To be honest, I query the advantages of doing it that way. A robot that has code to plan a smooth & optimal path around the track & calculates braking and steering accordingly will do much better (initially at least) than an AI robot that needs to learn this information. Perhaps bots that use a mix of the two (preplanning to begin with then learning to fine-tune any errors in the plan) would be the best solution.
In the year 74AD (the yearly calendar was adjusted in 2010, but took a few years to change globally, oh and AD now stands for Another Date), a man stumbles across a car from the past, but is baffled when it doesnt move by itself, and there a strange circle on one side of the car,.............anyways, havoc is created on the loads (formally roads, but people no longer know how to pronounce 'R's, and replaced them with 'L's) when this guy drives round at stupidly slow speeds, and the AI cars smash into walls and buildings, hahahhaha.
the car is a ferrari, taveling @ 160mph. normal cars now travel @ 160mpns (miles per nano second)
Wonder what will the do. Realise that you can flow the whole track allot faster
by cooperating with other cars or compete for every little gain (given that the objective
of the AI is to get around faster and not faster then others) and bog everything
down like most people on the roads today ?
It's sad that you are actually contemplating this so you can get drunk and have a computer
drive you home. Maybe you would also like a computer do most of the thinking for
you, especialy since you have so little time between all the things your daily life requeres
you to do. Maybe you just want too much and nothing will ever be enough.
And maybe this is the wrong place for a rant like this. But hey don't take it too hard ,
let the computers and tv do the thinking for you, and give you a nice answer that will
not disturb your conscience too much.
Does this mean I can finally leave my blowup doll at home and ride in the carpool lane?
but is it Linux-compatible?
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C1 bottles of beer on the wall. Take one down, pass it round... Oh, umm...
I'm all in favor of robot contests and all but more important, from my point of view, is the ability to share resources (such as test environments, robot chassis, sensors, vision-processing code, etc.) outside of the competition itself.
The biggest unnecessary impediment to robotic research right now, as I see it, is the difficulty researchers have in making comparisons between systems. You demonstrate your racing code on your robot in your test environment. I demo my code on my bot on my test track. The results are different but what does that show? On the other hand, if we both have the opportunity to try out our code on the same robot in the same test environment and mine clobbers yours, then the whole world can clearly see that mine works better. (Or, I suppose it is logically possible that yours would outperform mine, but that seems pretty unlikely, now doesn't it?)
We can get some kind of head-to-head comparisons in competitions, to be sure, but even then it is often just the environment that is the same. Typically the contestants are still providing all of their own hardware and software (as in the DARPA Grand Challenge and TFA). Even if we provide contestants the same hardware (xor the same software), limiting our comparison time to a couple of days a year impedes progress. We should be able to test our systems year 'round.
What we should be doing is making our code and our hardware and our test environments available to one another on a daily basis. If I want to see if I can evolve a better neurocontroller for your race car than you did, you should allow me to download my code onto your race car to drive around your track next month. Want to see if your code does a good job of driving my FIDO-class planetary rover over a simulated Martian surface? Download it onto my bot and run it in our Mars room or our outdoor OK/Mars test site. If you want to see if your rover hardware design can outperform the classic rocker-bogey design, pack it in a crate and ship it to us and we'll run it around our test environments for you.
Of course, it isn't quite as easy as that since the labs with the coolest robots in the world (which cost a pretty penny) can't spend all of their time and resources running experiments for other people at no cost - they have to get something out of the deal too. But that issue is not insurmountable.
I do applaud the provision of the simulation version of the race, which gets us running the same (simulated) hardware in the same (simulated) environment. (Interested readers should see http://julian.togelius.com/cig2007competition/ for the Java code. It is very simple and fun to try out.) The one question I have there, what is the license like for the simulator? I didn't see a README file or a note on the webpage. I didn't dig into individual source files or anything. Open source of some stripe would be nice, so that we can all improve it and share the improvements with one another.
Also, if anyone can suggest a more realistic racing simulation environment that could provide a better bridge to the real world competition that the simple 2D sim mentioned above, I'd appreciate it. An open interface is, of course, a must.
Dean
If the issue is the Rapture, I think most of the people left behind will have tougher aspects of their lives than whether their cars will function safely.
As far as testing goes, when we were working on the jeep two years ago, we had a nice, big, back yard to play, courtesy of as he has considerable acreage (20? 30? 40?). It cuts down additional neighbors and there's a lot of nature preserve. You can't hear the traffic a few blocks east of it from the major thoroughfare coming it's like a nature preserve in most of it.
When we were testing, it was someone in the front seat to stop it if something went wrong; also you had a walkie talkie to talk to the pit crew. Everyone had a two-hour shift. Big tips: Go before you start. Only drink water with you if you must. There's nothing like driving 100 yards away from the bathroom and you can't stop. You basically sat behind the wheel and if something looked abnormal, you could wire it back for feedback (or stop it if it seemed terrible).
The feature I've always thought would be nice for the open traffic vehicles is to be able to chain them together such that if ten cars are passing along a common stretch of distance, they could be spaced out, but a safe distance apart. It would be like a e-(train and caboose). The head of the chain would know when to stop for lines, to stop before railroad crossings, etc. That should reduce accidents, permit a decent speed beyond today, and perhaps give them a certain lane for e-trains and people would have to move over. Think of a rush-hour lane in some larger cities.
There would be the inherent problems of taking advantage (criminally). Broadcast a signal telling it to go ahead measurement ahead, then turn right, then stop. Ducks in a barrel. Pop a barricade up and watch the vehicle stop. Another sitting duck.
Most vehicles with fatalities to so because someone forgot there's a real-world Physics Police . They don't issue warnings.
As far as safety issues: at 4:12am this morning, someone hopped onto I-465 and started driving against traffic for nine miles. That is, until someone didn't react fast enough and they had a contest of car vs. (oncoming) SUV. All three died. The driver (at fault) couldn't switch directions (if they even realized it) when they realized there was a problem because the medians are made from nasty pieces of concrete. But they did have several exits to get off. They also could have pulled off to the shoulder or median and sought help. A lot of people managed to avoid the car and there was an onslaught of 911 calls.
Most of the good pictures aren't online, poorly uploaded, etc. The local affiliates showed cars which looked like the T-1000 when it was blasted to some pretty funky thin, rough edges. If they were still alive, there would have been nothing to pry apart because the pieces were just big enough to hold together but too thin to use the Jaws of Life against. I have seen a couple of these ((back in the day). A few with just one car: 90+mph and running into a street light. It hurt the light pole but goodly sized trees have a few, big scratches into the bark. Give yourself the task of scraping as much bark as you can in fifteen seconds. That will be considerately more damage than most accidents rip off.
Something which most people don't think about is: if the front end crumples too much, the engine has to go somewhere. It certainly doesn't jump out from under the hood and clear itself from the car. It's not going to deflate into a small piece of rubber. Imagine parking an engine in your passenger area at a pretty good speed and (people) trying to survive. Did you watch Twister and see the tower fly into the truck's front window? That's minor.
_____________________
As far as taking the keys away, my grandmother wi
I thought this would be time based trials, but read this "while outwitting the opponent cars but to do so it needs to be smart, it needs adapt to the behaviour of the other cars as it drives." Great, I get to spend my time developing a driving robot just to have another robot freak out and slam my car into the wall. Haven't they heard of baby steps? Hopefully, they'll use time trials and a run with a remote controlled car to weed out the obviously not ready ones. Seriously, if they just race then a quick start and swerving towards any other car will probably get you ahead since I think a lot of teams would program in collision avoidance.
There is an old kindergarten lesson that goes like this "Don't build anything taller than your head - it could come crashing down on you." That's what this reminds me of. You can build it. But is it wise to do so?
I think the important part would be to properly implement the congestion avoidance algorithms. So when the queues at an intersection get too long, we randomly select vehicles from the queue for instant destruction in a car crusher at the side of the road. Note that the contents of the packets, sorry cars, get deleted too. After a few weeks of this congestion will no longer be a problem.....
hmm, the captcha is cleanse...:)
... how does the rest of it go?
/dials Branson Information
Notice that in all the examples, the road is much wider than the car. That's not by accident.
Driving using reactive behaviors is easy if you have plenty of room. On narrow roads, though, those approaches fail. You have to look ahead. In fact, to drive in the real world, you need a controller that plots at least an S-curve ahead. Otherwise, you'll end up in a tight spot pointed in a direction that won't get you through.
You don't necessarily have to "plan", in the AI sense, but you need a fairly good dynamics prediction capability, after which you can run a reactive controller on the prediction.
We went through this with our DARPA Grand Challenge vehicle. We started out with a reactive planner, but it just couldn't deal with tight spots. Most of the other teams ended up with S-curve planners, too. The reason you need S-curves is that you need to be able to achieve both a desired position and direction at a point ahead of the vehicle. So you need a curve with at least two degrees of freedom.
The predictor needs to know enough about the vehicle dynamics to make reasonable predictions. For example, predicted S-curves have to be built knowing how fast you can change the steering angle and how tightly you can turn given the current speed and ground bank.
If you need to do this stuff, read up on adaptive model-based feedforward control. The idea is that you have a system that learns how the system behaves as the inputs change and builds a model. Inverting the model gives you a predictor. Given a predictor, you can control.
A useful feature of that approach is that, while you're using one predictor, you can be training a better one safely. Predictors are trained by watching; they don't have to be in control. So you can start out with some dumb controller and work your way up to better ones, without crashing. This is probably how mammals learn motor skills.
But I didn't see them, so I'll join the bandwagon:
In Soviet Russia, car drives you.
Those cars are definitely drifting! In that first video, all four "wheels" are clearly going sideways at the same time.
#naabhaprzrag, #sverubfr-000, #agi-fcbafberq, negvpyr[pynff*=' negvpyr-ary-'] { qvfcynl: abar !vzcbegnag; }
If a job's not worth doing, it's not worth doing right.
Safety. Yet another compelling reason to drive a mid-engined car.
It's hard to be religious when certain people are never incinerated by bolts of lightning.
Some posters have posited the "what if every car were an AI car" scenario. But the transformations all talk about changing our current cars into AI cars.
But I'd like to propose a different slant. If every car is an AI car, then how will this differ significantly from a distributed form of public transportation? If you break it down, the daily commute is filled with SUVs and a lone driver, with the SUV remaining parked (taking up space) for the whole day. So what about "transportion as a service" here (AKA: public transit and taxis)?
I mean, if I need to get to work and the wife needs to go shopping with the new-born, in many cases we need two cars. But if my car can drive itself home, then the wife can just wait the extra time and have her car back. Point is, we can optimize roadway usage, but we can also optimize car usage time. Communities could own car pools and "rent" them out. With communities co-ordinating their own commutes, but also with cars that can do intelligent pick-ups, you can get by with less cars and with less "big cars".
Think how much better the mileage will be without that 175 lbs of dead weight.
This is great news for road safety - now we just need to eliminate passengers as well as pedestrians and other road users so we can finally avoid people getting hurt in traffic.
Does anyone else feel like automating our current transportation is insanity compared to building a new transportation system that actually lends itself to automation?
Why are we trying so hard to make something designed to be operated by a human computerized so it stays on the road when we can make a road with rails on it?
You are checking your backups, aren't you?
In the overcrowded cities, the only thing that gets you from A to B is the "humanity" of other drivers, who let you take that left turn or entering the main street by slowing down, evan if they have priority. And many more situations that a robot would not understand, like anticipating a child's intention to cross the street, or some idiot's parking without signaling, and so on
Indeed, it could give you time to shave or whatever during the morning rush, but i wouldn't count on it
funny pics
You do it by Risk Analysis. It's easy to score impacts and compare the results. I understand the other answers, and they may have validity in some circumstances, but with Computer AI you are able to examine the situation, evaluate the expected outcome, and take whatever action produces the most appropriate of all the outcomes which are possible.
.....
.....
I would have no difficulty coding this up at all. I started my life coding Command and Control systems, and the Decision Support packages in these do exactly what I say above.
An added bonus is that all this data is available for the accident investigation. No longer do you need to say 'He jumped in front of me and I couldn't stop!'. You will be able to read off a printout looking like this:
Accident Report - 1109-22G/428-59B 12:08 1003473.209876GPS
1109-22G log readout
12:08:15.028 - Object moving from left on a collision course detected
12:08:15.028 - Full braking applied
12:08:15.029 - Override Broadcast to all objects within 40 yds of current position - details of proposed emergency manoeveure
12:08:15.030 - Security Handshake with oncoming vehicles established
12:08:15.031 - Object identified as running human - height 1.130m, mass 28.4 kg
12:08:15.032 - Oncoming vehicle 428-59B agrees minimum damage manoeveure
12:08:15.032 - Car trajectory adjustment begins to impact human on left crumple zone, oncoming vehicle 428-59B on rear right wing crumple zone
12:08:15.033 - Impact optimisation calculation begins
12:08:15.035 - Suspension height adjustment begins to optimise impact of human on left crumple zone
No option to walk? you should speak to your political representatives and get pedestrian walkways put in either side of the road. Even sub-Saharan African countries can afford to do this (see page 27) - it doesn't have to be expensive - so I guess your country needs to think about its priorities on spending. Which country are you writing from?
While on my daily -1 hr minimum each way- commute through northern va, en route to the DC metro area, I'm often
given to ponder what can be done to allievate this traffic. All the tools available, such as
rapid transit, light rail, all pale in the face of telecommuting. Since in the area surrounding DC at any rate,
no one actually DOES anything, they just drive, and sit in cubicles. They could do that at either home, or
general duty remote branch offices. All kinds of things could be done to allievate this more and more
horrific traffic, without adding a single square foot of asphalt.
But now, we can have more cars and less people driving!
What a great idea. That will really improve things.
Good thinking!
Keep up the good work.
Isn't this why we have mass transit in big cities? Actually, I think that it only makes sense when you're in a place where owning a car doesn't make any sense. Other than that, we need to invest in making better drivers in America. Coming from Louisiana and now living in Georgia, I can relate that that drivers in both places are TERRIBLE. Plus, there's the issue of when the technology goes wrong...you've seen the movies... Of course, this could all be like the argument that the old computer guys made when they said all we needed was maybe ten in the world or somesuch, and I could be COMPLETELY wrong and this is the Next Big Great Awesome Thing. However, we still need a place for people to drive...because it's FUN.
"Being legally classed as being blind, I would love to have a driverless car. I would live the independance it would bring."
;-)
Ever see Stevie Wonder's car?
Neither has he.
I'm just a simple mechanical engineer who's played around (written) a few programs that implement neural networks trained with genetic search algorithms, which is almost exactly what the article talks about. I'm also no expert on them, so take this for what it's worth.
Since this is /., most of you probably already know what genetic algorithms are and what neural networks are. For the two of you who don't, I'll give you the quick explanation: a neural network is function who's response to a set of inputs can be tuned by adjusting a set of numbers, and a genetic search algorithm is a stochastic optimization method you use when you have plenty of computing horsepower and absolutely no fucking clue how the system works.
The beauty of the genetic search algorithm (GSA) is also the fundamental reason this thing won't be driving cars anytime soon. GSA's optimize by experimentation, so they can and will come up with solutions that seem to work very efficiently, but make absolutely no sense to a human observer.
So, sure, you can have a GSA train a neural network for you that lets the car drive itself. With enough development time, it'll probably work incredibly well. But when the engineers sit down to figure out how it works, they'll have absolutely no idea, and they won't be able to deterministically audit the system. That's a bad thing when you move out of the carefully controlled environment of a computer simulation, or even the test tracks of an automotive proving ground.
What does the AI do when the left forward sensors see a stationary mid-sized red body, the right forward sensors see a small dark obstacle in the middle of the road, the lateral accelerometers are reading zero, and the right rear shock potentiometer shows the back of the car is in rebound? Well, unless your engineers can tell from the list of several million floating point numbers that comprise the neural network weights, the only way to tell is through experimentation. The AI might respond by doing nothing, it might swerve wildly, it might shut off the engine or flash the left turn signal. You've really got no way of knowing without trying it.
There's an inherent problem in deploying a technology that you can't audit. This problem is magnified by the fact that cars can kill people. This leaves you with the impossible task of trying to replicate every possible set of physical conditions on a test track. Until we have some way of doing that, no engineering group will release such technology for this particular purpose.
Traffic throughput is limited by how quickly you can get cars off a particular road. i.e. Parking. You can have cars 1/10th of a second apart, but if it takes 10 seconds to park the thing, or even 5 seconds to turn onto another road, that is the limiting factor.
If everyone has their own car, road performance is limited by parking bandwidth. Now, if everyone used a taxi...
Deleted
Is justs loves hows peoples pluralizes thes words towards.
--Richards
Practically all cars are designed to do this. The engine and transmission mounts break (nearly all of them consist of two metal plates connected to one another only by a piece of rubber vulcanized between them) and assuming the car is right side up, gravity will cause it to go under the car instead of ending up in your lap.
Of course, in basically all cases the engine is sitting on top of a cross member that connects the two sides of the front suspension, so in many types of collisions, this may or may not happen. But let's face it, life is made up of uncertainties and when you're moving down the road at a good clip, all kinds of crazy things are possible.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
Sigh.
Any technology distinguishable from magic is insufficiently advanced.
We make mistakes because we have to be flexible enough to handle the rather wild and chaotic world around us. That world isn't going to get any less chaotic, so AI systems are going to have to be similarly flexible, which means they're going to make mistakes in judgement.
There's no such thing as perfect reliability. Is 99% reliable enough? How many people are going to be killed by an AI at that level of reliability? How big are the lawsuits going to be at that level of reliability. How about 99.99% reliable? How many people per year is that? 10, 20? Again it comes down to the level of damages when an AI system kills someone, because it absolutely will happen, there will be lawsuits over the resulting deaths and the manufacturers will be held liable.
Deleted
Seriously, this isn't just for the elderly. Driving ceased to be fun for me long ago. If I had to do it only once a week on a nice stretch of fast highway, I might feel differently....
I hate to give a tedious, AOLer response, but... "Me too."
I HATE driving. Hate, hate, HATE it. Every day I put my life in the hands of the collective attentiveness of every idiot on the road out there, and spend over an hour in state somewhere between abject boredom and fear of impending death just to go to work and back. It's either break-neck, wild speeders or stop-and.. well, stop traffic where I live. It's absolutely terrible, and mass transit options vary between non-existent to impractical (like a 1-2 hour trip combined with a half-hour of walking). The sooner that I can give my car over to (and more importantly the sooner all those cell-phone using idiots can turn their SUVs over to) an AI, the better.
I miss being able to take public transit everywhere from my short three weeks in Japan years ago. It was mostly stress-free, and I could get some reading done instead of idling my brain while searching for danger on the road. One of these days, I hope to move somewhere where I can rely on mass transit again, but the way this country is going, auto-drive is more likely to be in my future than a good bus ride.
If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
In the 1939 Worlds Fair in New York, they predicted that everyone would have a flying car by 2000. Actually, most airliners can fly themselves from takeoff to landing.
The thing is, the grandparent is incorrect, the throughput isn't defined by how close cars are together... Sure, in theory on an infinitely long road which nobody ever wants to get off of it might be but in the real world where people park and turn off of main roads onto smaller ones, it's the interfaces which cause congestion and poor road throughput.
Quick thought experiment. We have an infintely long road, the cars are travelling 2 seconds apart. The throughput of the road is therefore about 1800 cars per hour. Now add a car park at the end of the road, it takes 10 seconds to get a ticket, open the barrier and enter the carpark. The road's throughput is now, 360 cars per hour because it's limited by the parking.
So... How long does it take to load a car onto a train? 2 minutes on a good day?
Say you... took the cars off the road. Created a simplified environment where there were no random incidents. No people on the road, no animals/children, no drunks, or nuts. Get rid of the steering wheel and turn the wheels into rollers on rails/guideway but kept the vehicles single person, as with cars. You would have a safe and reliable network which could go pretty much everywhere as with cars but one which wouldn't require AI, just a sensor in the front of the vehicle measuring relative speed & distance to the object in front. Now, if you did that and made stops offline, the throughput of the network could be defined by the distance between vehicles and in that simplified environment, computers could stop a vehicle very quickly indeed. Say 1/10th of a second between vehicles.
Deleted
...why we didn't develop something like the "cable car" system for personal vehicles.
Kind of like, a belly-hook (or whatever device) on new cars that could engage continuously-running sub-pavement (cables/chains/whatever they use) that are under all major highways.
Then you use your personal vehicle to drive around, but the minute you get on a major highway, you have the ability to link in, and just read a book or whatever. Since it's mechanical (no doubt supported by certain computer systems) the chances of failure could be minimal. Since all cars are moving at the same speed on a fixed path, even the consequences of failures wouldn't be too bad, I imagine.
That would maximize the benefits of personal vehicles (flexibility) with the energy-saving, pollution-reducing, and labor-saving benefits of mass transit over highway distances.
Personally, I'd love to take the bus/train more often, but too often REQUIRE having a car @ work for customer visits, unanticipated errands, etc. during the day.
(hoping some urban planner reads this, goes "EUREKA!" and hands me a giant contract to plan it.)
-Styopa
One of the neat things about automation is that it would have an amazing effect on the flow of traffic. Imagine if just 10% of the cars on the Interstates travelled at the speed limit? At first there would be more accidents as nuts try to road-rage a machine, but soon people would be forced to travel at the same speed. I think this alone would be beneficial to the traffic stats.
It'd be a Red Barchettaof course.
01101001 01100001 01101101 01101110 01101111 01110100 01100001 01101100 01100001 01110111 01111001 01100101 01110010
making better drivers is exactly what they're trying to do!
Technology is designed to help you, not replace you.
This is one of those evil and highly unethical technologies that I think we all should shun - like Nuclear Power/Bombs, Cloning, and anything else. People don't want their lives governed by machines. And yet, the slashdot crowd IS TOO FUCKING DENSE to ever realize this. It's a fucking farce.
Now that machines drive cars, we can all burn more fossil fuels and make the air we breath more cancerous and the weather more volatile for all of us. And the world will sink with the weight of the human race. Thank you, very much.
Not been to Boston but walked around part of Detroit with some local guys for a couple of days. Guess it must have been pretty central because they took me to a nice museum at one point. Seem to remember there were sidewalks there. Helsinki (Finland) pretty good for walking around as well: that is one city that's really nice to walk around, some great architecture. Not much need for a car there, trams are great as well.
I'd have to agree. I've ridden the city bus before, but I literally had a bus stop within a block of my house, and the destination (either to junior high or downtown) was also within a block.
Even so, I could compete with the time the bus took to get downtown from my place with my bicycle. And that was ~15 miles.
Sure, reading while riding wasn't bad, but once you have a car and insurance the bus doesn't look like as good of a deal. $1.25, the cost of an adult far, back then bought enough gas for 30 miles, or downtown and back. Total cost on the bus would be $2.50, or twice as much as the marginal cost on the car. Add in that the car is faster and leaves on your schedule, it's a tought competition.
That's why I like the idea of PRT. Eliminating all the stopping and even limited to 25mph, the individual car rail system would be able to beat both the car and the bus in speed, while allowing you to do something else, while being able to be powered by the relativly cheap electricity that's not being bought from a political hotbed.
I don't read AC A human right
Having vehicals moving at widley different sppeds causes accidents... well, really people who take risks to get around slower vehicals are the problem.
If everyon is going 85, and one person is doing 70 you will have the same problem. Yet no one is going below the speed limit.
The funny thing is, if everyone went the speed limit, most people would get to their destination faster, overall.
I love moving fast, but overall having a system what automates everyone speeds would be better.
The Kruger Dunning explains most post on