Slashdot Mirror
Computer Will Take On Formula 1 Champion
Jacky Baltes writes: "Thought that Deep Thought vs. Kasparov was a big deal. I am part of a research group that attempts to beat the world champion in Formula 1.
The goal of the Man v. Machine Challenge is to design and implement a
robotic system that can drive a F1 car faster than the current world champion.
You can have a look at the progress at the Man v. Machine Challenge Web site . We will had some more technical details about our control system design, data fusion, and car model to the site later.
So Michael, hold on to your head.
Jacky Baltes"
.. be able to make a phone call, shave and drink coffee? all while flipping off the guy behind it?
The Kruger Dunning explains most post on
It's not clear from the web site if this robotic car is to actually compete in real race conditions, or if they plan some farce where it's just doing speed laps solo? The first is a real challange, while the latter is a farce. It's a factory robot following a white line - only faster.
"How perfectly Goddamn delightful it all is, to be sure" Charles Crumb
but what I want to know is, can it parellel park?
Ciao.
PhatKat
This doesn't seem too much of a challange to me. I mean admittedly it's quite a feat of robotics and computer technology, but Race car driving is
mostly a test of agility and quick thinking isn't it? A computer with a good understanding of physics should be able to determine the perfect speeds and angles as well as determine when the tires are too bald and extrapolate when more gas is needed etc...
Deep blue actually had to outhink a human(If you can call Kasparov human:) ) without just simply being faster. Even deep blue couldn't know all the possible chess moves. Although quick thinking was certainly a part of it, it seems more than that.
Not to knock what you're doing. The technology just in the robotics to controll the car must be amazing, but it might be better to compare the test to the first cars that could outrun a horse rather than Deep blue.
"as plurdled gabbleblotchits on a lurgid bee" - Prostetnic Vogon Jeltz. (One man's humorous is another mans flamebait)
Thank goodness its not nascar, Blech!
Buying a Dell computer is equivalent to dropping the soap in a prison shower.
Computer cars don't fear making a miscalculation that will kill them. If the human driver had no way of dying, then it'd be a bit more of a fair fight.
Sorry had to say it, Imagine a Beowulf Cluster of those...
This seems to me like the first steps of an AI which could be dangerous. Don't get me wrong, while creating an AI is a great leap for mankind could it also be the downfall of mankind? Also Will this "robot" be driving against one person or will it be driving in a larger race? I think for it to be a true test of man vs. machine it should be in a real race, having to deal with multiple opponents, as men are forced to do.
When a man sits with a pretty girl for an hour, it seems like a minute. But let him sit on a hot stove for a minute-- and it's longer than any hour. That's relativity.
-- Albert Einstein
May be you should start with RARS first.
Is the race for the best time or is it a head-to-head race? Can't tell since the web site isn't responding.
If the later, I wouldn't want to be the human racer. The current state of autonomous computer controlled driving is pretty lousy. One miscalculation could result in an accident. Also, human drivers are a little more cautious with risky moves since the consequences could mean death or injury. Will this computer controlled driver be cautious about the human life of the other driver when considering grazing the other car in passing?
pronoblem
... that nobody's made a Microsoft joke about the car literally crashing on Windows 2000.
c'mon, people.
J
sorry couldn't resist.
Really though F1 cars are fast and dangerous, I really hope they do a good job in the design.
For instance what is going to happen if a tire blows out, sensor/circuit fails.
They might do okay, but I would be worried about low cost implementations coming onto the road too soon.
I can see it now, AI-Andi the T-Shirt, AI-Andi the matched luggage, AI-Andi the toilet paper, and finally AI-Andi the flame thrower.
The best way to accelerate a windows box is at 9.8 meters per second square.
The human driver is going to win, but after he wins he is going to stumble out of his car, sweating, and die of exhaustion.
Hopefully I didn't put any [] around my words.
They fed the program Kasparov's entire game history while keeping its game history secret from Kasparov. Normally in competitive chess you are allowed to study your opponents past games in order to learn what tactics they are likely to use. In this case, Kasparov wasn't allowed to do that. The fact that he went ahead with the game anyway was probably due to overconfidence on his part.
The folks at IBM seemed to realize that they won merely because of the setup, and thus when challenged for a rematch by Kasparov, they said they weren't interested, because they had "done everything they set out to do". (Personally, I think they were scared they would loose in a fair match.)
Kasparov has stated publicly that if the "Deep Blue" team actually abided by the rules of competitive chess, he will "tear [Deep Blue] to pieces". They have so far declined.
That said, even if the machine were able to beat the best human player in a fair match, it still would not be that remarkable, because computer chess programs are still limited to the "brute force" approach, where they pick their next move by simply searching as big an area of the total possible game tree as possible. The human mind does it differently, only examining at most a dozen or so possible moves before deciding. Ho the brain can pick such strategic moves without searching a significant portion of the total possible game tree is still one of the great mysteries of cognitive science.
Free Hans!
Consider chess: you have a vast archive of previous games, a relatively simple domain, the ability to test millions of boards a second, almost free live testing, and almost no financial penalty for mistakes. Contrast F1 racing: no archive, complex domain, almost no simulation ability, real testing costs $1000/hr, the mistake penalty is $100,000.
This is either hopelessly naive or a scam: after three years, you might get an AI around the track at 100MPH. Judging from the website, it's a scam: they talk about all the great value of the webhits and PR, ask for sponsors, etc. There is almost no info on the AI approach, etc.
Looks like nothing but a money sink to me.
The research team should contact the Knight Industries or the Foundation for Law and Government, and hire Bonnie as soon as possible.
(Yes, I read this post and thought, "Jesus Christ, make a Knight Rider reference as soon as possible." May others come and do it better.)
We may never see a robot piloted race car on the same track as the human driven cars, but I could see competing robots racing against each other. Then, it becomes more a matter of which team has the best programmers, as well as the best pit crew, etc.
needs a little r2d2 unit mounted on the top
Deep Thought is the name of the machine created by the mice in Douglas Adams' Hitchhiker's Guide to the Galaxy, that determined the ultimate answer to life, the universe and everything.
Deep Blue is the name of the IBM machine that beat Kasparov the second time they played.
On a semi-related subject...
Does anyone know anything about solid state gyroscopes? I read about a gyro once that worked using a length of fiber optic cable wound into a loop. You passed a laser through a splitter, one beam down the cable, and then both beams hitting a light sensor. When you spun the loop, it would cause the interference pattern to change, thus measuring rotation. The dynamic range of its operation was astounding, like from 1000RPM down to .0001 RPM or something crazy.
From a project like this to succeed, I would imagine you would need something like this that could withstand the G forces while giving you extremely accurate results.
It seemed so simple that I figured they would have taken over the world by now with a million uses, but I haven't heard anything since. Anyone know anything?
--
Sometimes it's best to just let stupid people be stupid.
I'd rather these guys did something worthy.
I'd pay good money if some robotics genius could invent a machine that I could hook up to a standard Tornado foosball table that could beat me in Foosball. I'll pitch in some starter money if anyone cares to help on starting a challenge for a FOOSBALL ROBOT vs MAN contest.
It's not really about beating the human opponent - it's about making a bigger, nastier computer. The human is just a benchmark, and not a very good one at that because we tent to be pretty inconsistent.
The Deep Blue-Kasparov fight was lost to begin with because Deep Blue could see a dozen or so moves ahead for any given board configuration, elimiate the ones that it was programmed to think unlikely and then pick the one that left it in the best situation given a set of rules. People don't do that, at least not on the scale that a computer can, not to mention the mistakes we make, so Kasparov was doomed to loose eventually, if not to Deep Blue then to Really Deep Blue. It was all about how quickly and how well the computer could "solve" the given board configuration.
This race is no different. It will be a lot more challenging because the inputs are infinitely more complex than in chess, and the proper course of action is sometimes not clearly defined, but it will just be a horribly complex formula of some sort that tells the car how fast to go. With other opponents on the track, the level of complexity goes up, but it's still just a formula.
Me.Speed = NewSpeed ( frTireTemp(), flTireTemp(), rrTireTemp(), rlTireTemp(), frTirePSI(),... )
What I'd like to see is Deep Blue explain why the chicken crossed the road or what's the ultimate question to the ultimate answer, or to just drink 4 pints of beer and try to pick up...
----------
Even as you read this, your pants are strangling your loins! Aaa!
No, but perhaps against Schumacher, Häkkinen or Villeneuve?
If we are talking about a real race situation and not a lame solo time-trial, the robots could easily win.
How? By teaming up. They are better equipped to handle the tiny calibrations in track condition in order to maintain a constant speed.
All the robots would have to do is to get 3 or 4 robot cars to surround a human driver on 3 sides, then slowly force him into the wall. Repeat x # of human drivers.
Of course, it would be more amusing if the robots just followed a hive mentality. Then they could just assign one robot car as the "Queen" and have all of the "workers" suicide into human cars. At a cost of 1 robot per 1 human, the machines would see that as a positive math situation. And they could probably also engineer situations to take out multiple humans with just one robot car.
Now THAT would be good entertainment...
------
Let me give you the lowdown
But a computer has no such fear. It makes decisions based on programming. So let's say that it cuts too close to another car for whatever reason, and in the collision the driver of the other car dies. Is the programmer liable? After all, he is the one who effectively made the decision to cut that close to the other car. But it doesn't affect him negatively because he was never at risk. So someone is going to sue him, saying that he was careless because he was never at risk.
I wouldn't touch this project with a 10-foot pole. If this car ever drives on a real track, it's going to end badly.
--
And the men who hold high places must be the ones who start
To mold a new reality... closer to the heart
Same FIA Formula 1 car for both operators
Computer must mechanically operate the same controls used by the human;
Computer must fit in the same space as the human, including power source.
Computer can have no electrical connection to vehicle for power or sensors; all its sensors must be self contained, and have no physical extension beyond what is allowed the human (no camera through the floor to follow the line :-)
Computer can have as many hands, legs, arms and eyes as it likes
Computer gets human equivalent sensors only - visible light vision and accelerometers; no radar, sonar, or active illumination allowed.
Computer controlled car must meet same weight requirements as the human/car combination
I won't demand race/traffic interactions. Solo qualification laps will suffice. Even under these conditions, I'll take the human for ten years easily, and probably twenty years.
I don't think it's an "AI" problem as much as a robotics, sensor, and machine vision problem. I don't think there's been so much progress in the last 20 years that it's feasible.
Though I sure Frank Williams and Patrick Head would sign it up as soon as it was available.
-dB
"It if was easy to do, we'd find someone cheaper than you to do it."
Seriously, what's the point of having a computer do the same type of repetitive tasks (like race car driving) when we already know the computer will eventually win.
Computers don't get tired, they don't get thirsty, and they don't have to urinate so it's hardly a fair challenge.
...people would stop watching. I highly doubt robotic races will ever garner much more attention than the geek audience. They may get an initial first look from race fans, but don't tell me that 50% of the rednecks at the Daytona 500 wouldn't quit watching if the cars stopped crashing.
On a side note, remote controlled cars, without the fear of the driver losing his life, might be something else entirely. I'd love to see an Indy car throw a piston on the straightaway because the driver was trying to push it to 250mph!
Okay, the webpage says they'll race the World Champion (whoever that will be in three years I guess). I wonder if they stopped to think about whether someone like Schumacher or Villaneuve (please forgive possible misspellings) would even consider racing against a machine? I know I wouldn't, and the racers I know would definately scoff at the idea.
Second thought, I wonder why they chose F1? It wasn't that long ago that the governing body of F1 started being more strict on what kinds of electronics could be in the cars. Someone said (can't recall who) something about the cars almost being smart enough to drive themselves. If they were truly out to show that the computer can do it better, they should put it in a World of Outlaws car (struggle to go straight, let go to go around the turn, sideways). Or how about Motorcross (the old kind, where you weren't allowed to put your feet down)? Anybody watched "On Any Sunday"? Let's see a robot try to put a motorcycle over a 30-ish inch log laying across the trail, and then make the 90 degree turn right after the log... Most people can't do that..
Del
... it's already been done: Schuey is a machine!
The nice thing about Formula ford is that there are large numbers of near-identical cars available off-the-shelf for a realistic pricetag, yet the skills required to drive a formula ford well are very, very similar to an F1 car.
Any sufficiently advanced technology is indistinguishable from a rigged demo
--Andy Finkel (J. Klass?)
- Note that the first items in their menu, and a large portion of the site, are about sponsorship, how big of an opportunity for advertising this is, etc. etc. I've never heard of a research project with a "Supporters' club".
- As has been mentioned, it wasn't Deep Thought that "beat" Kasparov (it's argued that it was hardly a fair fight), it was Deep Blue. Methinks someone who was studying AI techniques and relevant prior AI work would know that.
- The few technical parts read like a sophomore year computer engineering project proposal. high on fluff, low on details.
- Nowhere does it give any details of what this challenge really is. Is it just negotiating a car around an oval track at high speed? Is it a real race with other drivers? The real goal is unclear.
I think it's a neat idea and if it's for real, hey, I applaud them. Personally, I take it with a grain of salt the size of Texas. Sounds like someone trying to grab attention, maybe go IPO, get-rich-quick. The whole thing sounds like a sales pitch, and I'm not buying."This message is composed of 100% recycled electrons."
Any sufficiently advanced technology is indistinguishable from a rigged demo
--Andy Finkel (J. Klass?)
ferrari spent over a billion dollars on it's f1 team this year. Michael Shumaker is the highest paid athleate at 50 million a year. How much do u plan on spending?
Of course.
Computer must mechanically operate the same controls used by the human;
Why? The extra mechanics involved adds no real difficulty to the computers task. I agree that the computer should not have any extra control over the car, but what advantage does the computer get by issuing the set_brake_level(50); command insted of extend_left_foot(50); ?
Computer must fit in the same space as the human, including power source.
Yes, we want identical cars, but I'd rather state "Computer must be contained in the car. No remote control." That said I'd be equally impressed by a remote control setup as long as all sensors were in the car
Computer can have no electrical connection to vehicle for power or sensors; all its sensors must be self contained, and have no physical extension beyond what is allowed the human (no camera through the floor to follow the line :-)
Same as for the controls. Wether the computer has a direct feed from the cars sensors or points a camera at the dials and does some image processing is not important. What matters is that the computer must not have access to more information about vehicle status than the human.
Computer can have as many hands, legs, arms and eyes as it likes
Yes, but as I said: I prefer a display of AI not robitics.
Computer gets human equivalent sensors only - visible light vision and accelerometers; no radar, sonar, or active illumination allowed.
Hart to tell what is equivalent. A human has stereoscopic vision to measure distances. I think that would be a hard task to match.
Computer controlled car must meet same weight requirements as the human/car combination
Definitely. The question will rather be: Should the human car add ballast to make up for a heavy computer? I don't think it will be very lightweight...
*IF* this ever comes to be, I'd like to scale it down to an AI problem as much as possible, not a robotics/sensor contest. "Can a computer drive a car at 300 km/h?" is a much more interesting question than "Can a robot controlled by that computer operate the car?"
All opinions are my own - until criticized
Remember Niki Lauder...'robot','machine'. The names designed when Niki won a race after continually lapping past an overturned car in which the driver was killed. (it was unfair of course...tragic circumstance involved). Any arrogant engineers who really think they could pull this stunt off, or even believe that others could do it, deserve no more than the half-arsed web page they've already got.
But suppose the projects definition of "beating a human driver" were to mean only "getting a car around a race track faster than the wetware controller can." If the "race" were separate time trials (i.e. on an empty track), then this prject is not as totally insane as it would appear at first glance.
I don't see any mention of them running the race together, with lots of drivers. My guess is they are wankers who programmed a robot to drive an empty course the fastest they could with a formula one car. Given the weight reduction and the pretty close repeatability it may be able to beat the driver eventually this way. If they ran *together* then the other driver could certainly use his knowledge of passing and cutting the robot off to great advantage.
-Ben
The picture on the site is of a recent F1 start.
A computer should be able to get a better lap time than a human, but that's not the same as winning in F1. The technology of the cars has a lot to do with winning. (Look at what happened to Irvine when he left Ferrari for Ford!) Driving in traffic would probably be beyond the computer.
A question that I would have is: who would do better in an unusual situation? Say the car starts spinning, or something on the car isn't working 100%?
--
Sadly that's not going to happen, mainly because he died in 1992. But his stories and books probably did inspire a lot of the people working in robotics to enter the field in the first place.
"How perfectly Goddamn delightful it all is, to be sure" Charles Crumb
There was one Knight Rider episode that featured Michael's evin twin. The evil twin drove an evil robot semi. What was the evil twin's name? I think it was something like Garth. What was the name of the evil truck?
cpeterso
I've just gotta say, wow, Slashdot has changed - two years ago, you could hardly find any motorheads here and now look at us talk about complex F1 traction control and engine management systems. I love it!
As someone who originally trained as a robotics engineer and has spent enough time on the track to judge the difficulty of the problem, my money is on the "meat puppets" for the next 50 years or so... (That's 30 years for procedural AI to finally die and 20 years to actually solve the problem.) But no computer will ever be likely to drive like Nuvolari, Fangio, or Foyt, as that kind of performance is beyond technical and requires soul.
"The future's good and the present is nothing to sneeze at." - Roblimo's last
Why do all these AI researchers keep wasting their time on toy projects like games and driving cars? Do people no like playing the games or driving the cars themselves? What we need are computers that can handle the dangerous or undesirable jobs that people would rather not do, like policing high crime areas. Or dull stuff that isn't worth paying a full time employee for, like a night watchman. Just think how much safer we would all be if every business had a heavily armed security droid at the door.