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"
I just want to see the successive videos from the "earlier stages" of the system's development (ie I wanna see the computer driven cars eat it)
I am joe american
This whole saga kinda reminds us all of the whole Hyperion thing (Jim Clark's 200 foot sailboat, sailed by 25 sgi workstations) - now that's a project.
-f
.. 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)
The computer that took on Kasparov was Deep Blue. It was not named Deep Thought as the posting would indicate.
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
Deep Thought? Where's a good Jack Handey quip when you need it?
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?
soon the problem with riding in a cab won't be the question whether the cab driver can speak english, but what operating system does the cab run.
--- d'oh
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.
The computer has come a long way... and it steadly is replacing everything we have come to know of. It advances and improves, getting to be able to do everything. This is simply further proof of the perfectness of the computer, a perfectness... the ultimate goals of a computer, and we are simply making them perfect. Is that our purpose, to make computers perfect so that they will help us incredibly in the future? I fear that as beings become more perfect, the imperfect are decided to be exterminated, that means us, the humans. The comptuer is a dangerous tool and we see every day how it is getting stronger and stronger, we are undermining ourselves...
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!
Either they don't know the meaning of "Pandora's Box", or the reports I read about WinCE to be used in cars were true. Oh, the humanity!
.-.
Help savingAmigaOS and a free PowerPC market
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.)
pronoblem
Would you prefer your car to blue-screen-of-death, or to dump core?
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.
I think that this challenge is somehow dumb. It won't be long till some genius claims he can build a robot that can run faster than any human, or even do a better job at adding 312893 numbers with 7942 digits each.
I'm an American, and as such, absolutely fucking LOVE F1.
Pink-boy.
The revolution will NOT be televised.
GM had cars under computer control making competitive laps back in the late 60s when they were working with Roger Penske and Paul Van Valkenburg on the Trans Am Camaros.
Running a fast lap and winning a race are two very different kettles of fish.
It has been noted that nearly ALL drivers of F1 machinery can turn nearly equal laps in equal cars, but only a half dozen tops are capable of winning races. At this level of expertise and competition pschology is the detirming factor.
You want to race a bot against Shumacher, Hakkinen or Villinueve?
I'll bet on the human driver every damn time.
The site isn't really responding so I can't find out the answer. But what is the robot going to drive? I mean we are talking about cars that cost millions of dollars, and then to develop a car...that cost hundreds of millions. I don't think a team is going to let a robot "borrow" their car...plus in order to have a hope it would have to be either Ferarri or McLaren/Mercades...and I don't think Ferarri want a robot to beat there 20-30 million a year driver...and McLaren/Mercades doesn't want a robot to beat a driver Hakkinen couldn' t in there car.
First the robot has to weigh less than 100 kilos, and have enough power to last > 70 laps. The driver has to sit in the car and drive- after all this is F1, not remote control race.
It doesn't matter that much if the robot is super light either, because the F1 rules states that you need to add ballast to make up a certain weight for the driver. However having the robot light means you can put the ballast in better places.
Second the robot has to qualify with lots of other people zipping around getting in the way.
Then if it doesn't get pole position it has to pass a driver or two who may be working together to stop it.
When you are overtaking someone ahead of you, that person is not going to make it easy - you have to pick just the right time when that person will let you pass. Can the computer figure that out?
When you are lapping, you can wait - most backmarkers will let you pass, but some have their own battles. And you must beware the recent Hakkinen manouevre!
Some people have mentioned that the computer doesn't mind dying. But what they don't realize is that if you drive into another car, it's very likely BOTH of you are out of the race. And if they can prove that you have been negligent or were driving very dangerously you could be fined or banned.
If you smash into other cars too often, your team goes out of business - no more money to fix up your car.
These are real life stuff, which the computer cannot avoid. And computers aren't so good in the messy world.
Personally, I think they should prove themselves out with kart racing first. If they can't do kart racing, then forget about F1. Don't waste time and money.
I bet I can learn to drive real fast on an empty track. But when you dump other drivers and variable conditions into the picture it gets a LOT harder.
Cheerio,
Link.
I watch F1 regularly, and I'd say the challenge of controlling a 900HP 1300lb car on any given track to get a better time than a human would be quite a feat. These guys fly over curbs like crazy and then just correct like a madman; I can't see a robot doing it. Of course, they have unbelievable data sets to go on, so it's kind of like reverse-engineering the thing.
But I'd hate to see what it does when it spins in the rain, or clouts a tire on the car in front of it. Yikes. Keep a finger on the kill switch, fer sure.
The revolution will NOT be televised.
Competing in physical activities is no big deal even though we have to keep in mind the twist and turns the car has to take to avoid other racers, but here again there is no risk to life to the experimental car. Machines have since ages been doing things more efficiently. All through ages we have known that man was not the sturdiest nor the fastest, it is just one species of the planet. But this just one species has made the world go round it and no car racing machine can change that. For the believers here's a line --even though we may achieve great things we are still made by God and he is superior, so also we made machines and we are the gods to them and interfere in their lives as to our whim and fancy.
There's always sufficient, but not always at the right place nor for the right folks.
needs a little r2d2 unit mounted on the top
I'd hate to be the one paying for alpha testing. That might get expensive. Especially if they use Windows. Image a crash to cost hundreds of thousands of dollars..
-- "Perceptions create reality. By changing your perceptions you change your reality."
If they can design a perfect race car program, imagine a perfect flight combat program. (I get shot down if I don't play on EASY level...) There'll be pilotless drones shooting down even the best Top Guns and jet-jocks. I know reconaisence drones have been around since Viet Nam. I even have a theory that the stealth fighter that was shot down in Serbia never even HAD a pilot. Those things are programmed to land on their own if the pilot is unconscious, why couldn't it fly on its own too?
The top drivers know how to help the engineers set up the car better.
:). All the best drivers have managed to win races even with faulty or inferior cars.
After a few laps, they're smart enough to tell the engineers what sort of setup they need - suspension, aerodynamics etc.
And let's see the robot drive with some gears missing and win
Link.
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.
Deep Thought was the computer in the Hitchhikers Guide to the Galaxy that took 8 billion years to calculate the meaning of life. 7x7=42 Earth was built as a more powerful computer to replace Deep Thought.
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!
Really, any Machine-vs-Man challenge is in reality A-Crowd-Of-Thinkers-vs-one-man. Was it Kasparov, will it ongoing F1. And Russian poslovitsa sais: vmeste i batku bit' veselej (It is funnier to beat the dad together).
A bunch of people almost always smarter than one man, so nothing interesting (to me).
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
I doubt that the computer that will drive the car will actually be in the car. A computer powerful enough to do this today would need quite a large chunk of space for both itself and the small army of people that will be needed to do the voodoo dance around it to make it work. It would probably not even be next to the track but instead sitting somewhere dry and cold (Canada? ;) and control the car by remote. Even if the computer could be made to fit, they probably wouldn't put it in the car anyway - what do you tell the press when your billion-dollar computer hits the wall on the first turn of its first run at 200mph and becomes one with the soil?
----------
Even as you read this, your pants are strangling your loins! Aaa!
F1 racing is more complex than it seems, with much strategy involved, and the science behind automobile racing isn't even that well known in an exact quantitative sense. It's more of an art than science. I'd guess the computer-driven car will either crash or get passed by human drivers at every corner. You can't even get a computer to keep a bus to stay in a lane on a highway reliably.
Ok, that's about enough... The tiny robots produced by a robot crawling around the room is ok. You can stomp them if they attempt anything (take on the world). But if you make a computer/robot that seats in a vheicle that runs at 300kmph and program it to compete with humans - Please be carefull with the programming. As the previous poster mentioned, a beowolf cluster of these gone on a rampage ("we are the robots - we have to win against the humans - we have cars that can make swedish meatballs of all pedestrians.......") would NOT be my idea of fun!.. Please remember the laws of robotics (Assimov) and bur them on a unflashable and unremovable ROM!!!
--
"No se rinde el gallo rojo, sólo cuando ya está muerto."
$HOME is where the
-- silver_p
I've seen enough of those web sites that come out with some amazing idea but with nothing in their hands but an HTM (as in .htm) editor. It used to be websites about new OS projects.. now something different for a change.s e ? ..it sounds very complex but it's more viable than crashing F1 cars.
..bha.. I just think this is just a scam.. they are already talking about fame and magazines !
I can see already one main reason why this is just fiction:
Who is going to give an F1 car to train someone's neural network/genetic algorithm/whatever-they-don't-tell-us-they-will-u
You can make a $1000 AIBO fall a thousand times while training its gaits but training a 10 million dollars car at 300kph is a different thing.
Before even thinking about F1 cars they should try to build enough AI to reliably drive a standard car without getting out of the track. Mercedes R&D recently did that and it was quite an a chievement.
I think they should rather concentrate on extremely accurate simulations to first be able to reproduce the a car on a track (things that are actually being done in F1 nowdays) and then try to plug an AI into the simulation. Maybe even simulate the vision system..
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.
I thought of an episode of Cyber Formula Sin when there was this problem with the Asurada? It was pretty cool. While the pit crew fixed the car, the programmer (hot chick, typical anime ;-) started Windows (?) and started editing the C program of the car (in a DOS prompt of course...)
...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?)
They speak of the robot driving the car as a passenger. There are onboard computer systems in cadillacs today that can prevent you from breaking the car loose from the road. In fact, they do everything short of turning the wheel for you. If I was to develop a computer controlled race car, I would use telemetry, lateral acceleration sensors, and other sensors to keep the car on the edge of control. The race car driver's job today is to get the car, in it's static configuration, around the track as fast as possible. The computer controlled car has the ability to change damping rates, roll rates, and perform yaw control on the car, things that a human driver cannot. In fact, all the 'driver' has to do is point the car in the direction of the corners and hold the accelerator to the floor, the computer will do the rest.
Don't believe everything you read.
- 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."
You're sorta cute.
to see how the computer does the tear-off's. Will they have a computer generated voice do Murray Walker too? In a real Brit accent??
there are no stupid questions, but there are a lot of inquisitive idiots
Hey, German is my native language!
While the premise certainly is interesting, I doubt these guys will get anywhere. The site appears to be run more by marketing types than techies. They've already devoted more web space to the marketing/advertising opportunities than they have to any of the technical issues. I'm sure they've already started focus group testing...
"The question of whether a computer can think is no more interesting than that of whether a submarine can swim" -EWD
The founder resumes look great - but the rest of the site - well, read:
So they're founding a pr/marketing company. And they're setting up a separate R+D group - composed entirely of Other People's Money. (Not to mention other people's technology and manpower.)So it all seems a little off to me. But then again, they wouldn't be the first to run a car on vapor.
Reading the summaries, especially the research section. The persons involved obviously know something about driving, but know next to nothing about how to practically apply AI to it. Also traing problems. The quote:
really cracks me up. I guess they expect to crash lots and lots of cars (or maybe they're going to build their own tracks with all of the money people are going to throw at them).They also seem to be treating "AI" as this big black box not dividing up the task levels as I would expect: high-level strategy ("Tire/Gas use strategy, push him now, hold back etc."); lap-by-lap decision making ("On my last lap the car felt kind of loose in this corner, I should take it easy in this corner"), and instantaneous decision making ("Rear wheels slipping unexpectedly, steering response").
The vague use of the phrases like "store all actions" shows way too much naivete for this to be a serious project. Seeing how they seem to be leaning towards "neural nets" as the AI implementation of choice, (and considering all of the training+experience it takes to become a halfway decent F1 driver), they're going to need lots of cars/computers/gps sensors/gyros/money.
I think they would be alot better off trying to build a stock car that could complete a circuit by itself in decent time. From there maybe they could build up to multi-car go-cart racing. And in 10 years, who knows? 10 years ago 10MB RAM was huge and 50MHz was inconceivable.
Verdict: If they're serious, and try to do F1 from the start, the startup $10 million should disappear right quick, never to be seen again.
There is no trap so deadly as the trap you set for yourself
There is no trap so deadly as the trap you set for yourself
-Raymond Chandler, The Long Goodbye
Fluffy technical details.
Two guys, one with only entrepreneurial interest.
Overemphasis on promotion and marketing.
No pictures of current or previous hardware/software.
Compare with the Thrust SSC site . Still going strong three years after the feat.
If you don't want to repeat the past, stop living in it.
How would the robot pop open the champaign and soke every one around it?
ABS only brakes better on uneven/extreme slippage situations. If your ABS is going off on dry pavement, you aren't riding the edge of friction, you've passed it. Any F1 racer that tried to use traction control would get lapped by the 3rd lap. Ever seriously watched F1 race? Their tires slip all of the time, and alot of the time, it's deliberate. ABS and traction control are trying to minimize slip. Drivers are trying to go fast.
There is no trap so deadly as the trap you set for yourself
There is no trap so deadly as the trap you set for yourself
-Raymond Chandler, The Long Goodbye
Server connection dropped
Reestablishing connection...
logging in...
crash!
There is no trap so deadly as the trap you set for yourself
There is no trap so deadly as the trap you set for yourself
-Raymond Chandler, The Long Goodbye
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?
(no females drivers in F1)
.oO0Oo.
I remember 1 on the starting grid once
There are places where the networks are not touching,and there are places where they are-Boeing's Lori Gunter
Given the extremely complex nature of Formula One cars and their development, this would seem to be a very human oriented competition. In this race I think the computer driver will have very very little to do with who wins. How can a group come out of nowhere and beat Ferrari at their own game?
One solution would be equal cars. This gives the human an advantage, because it takes away the computer's main advantage: it can be placed anywhere on/in the car, and thus the car can assume a more efficient shape.
A highly mechanical competition like this just isn't much of a game. Something more like tennis or chess is more fair - a superior machine is not enough to win.
Still, it is an interesting idea, as long as I don't have to pay for it.
--Jeff
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
Go fast, turn left
Sort of like watching the tumble dryer.
"Go blue underwear, GO!"
"Oh my god NO! my left sock crashed into a T-shirt!"
All opinions are my own - until criticized
Yup. I think we have an interesting convergence on the way here. i notice that automation like this is steadly becoming more practical and reliable, while at the same time Network Television (the world over) is now or is soon going to be desperate for 'blockbuster' productions (survivor, various millionaire shows, etc).
The Network TV shows of the future will need to be more outrageous than anything we've seen today, this has been the trend since the introduction of television. The networks will embrace automated sports sooner or later, and then all the networks will begin the cycle of one-upping one another.
Networks will arrange increasingly more lavish sponsorships for the teams. Increasingly more sophisticated robotics will be employed. This will spur development of more sophisticated robotics, providing a financial incentive, even. Amazingly, marketing dollars will be converted to applied engineering for once!
Surely, some day soon we'll see a humanoid deathmatch robot game with weapons like pick-axes and swords and maces. It just seems obvious that as soon as honda or sony releases a reliable, well designed humoid robot, someone somewhere will arrange a fight to the death between two of em.
This would eventually develop into a network style 'Robot Joust' series. perhaps the robotics will even take the form of knights in full plate? I can see it now, "The Black Knight vs. The Golem!", brought to you by Napster(tm)(maybe we could even see a writhing clockwork dragon? as long as it doesn't violate the weight limit...). The remote control controls would be similar to a marionette, running on 900mhz cordless phones. This is a sport that hobbyists will be able to afford (and much more satisfying than the SCA people!), weekend legues will spring up, joy will be had.
As the other networks champ at the bit, we'll see a bidding war errupt over Quake and UT licenses, for the purpose of adapting them to live TV. the combatants would all be mechanical, and maybe ther'll be a a limit on the number of bots you're able to enter, say 10. the course will be in a bunker, at the bottom of an old strip mine, covered in sand and gravel. Ther'll be cameras everywhere, literally covering every singele vantage, and at frame rates in the 1000's, like in that
If no adequate AI were developed (imagine a $150,000 robot bouncing endlessly into a wall) or if completely automated combatants are unpopular, a human 'driver' would be used. this seems the most satisfying and attaractive for viewers IMHO. This driver would use a regulation 'terminal', a specialised FPS environment. very deep immersion, but nothing like direct machine-brain interfacing or anything. And thus would lead to the debut of the world's first true Robotic [greasy haired, pimply faced, overweight] Sports Hero.
But there's a problem. It isn't much fun watching UT on TV, because by this late date, watching live footage of actual robots is precisely indestinguishable from playing it yourself. Remember, we're talking a couple console generations down the line, playstation 5 stuff. The ratings will flail, and the networks will fall back to torturing Average Joe, or maybe brewing up a war to cover.
:)Fudboy
:)Fudboy
I guess I'm only a Fudboy, looking for that real Transmeta
A number of people have suggested that this might be a hoax. I can't say absolutely that that isn't the case, but Dr. Baltes is a senior lecturer at Auckland University which I attend.
This project has been mentioned in my classes - I understand a number of staff are involved from the robotics and computer vision groups.
It certainly sounds incredible, but I'd say it's true.
They might not be able to match Schumacher in terms of speed, but I bet they alrady have a computer with more carisma :)
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.
Having the car automatically fix itself and compensate for problems when things go wrong is yet another really interesting technical problem. Having a reliable and consistent robot driver would make it a lot safer and more practical to test such solutions.
For instance, suppose the car had a way to know your Ford Exploder had had a blowout and automatically adjusted the steering to prevent the car from flipping over. With a robot driver around, you could fine-tune the sensor and the corrective handling routines without risk of loss of life.
Oh, and then we'd probably want to add a VUI (voice user interface) that responds to requests such as "One of the stabilizers has broken loose; see if you can lock it down!" :-)
I play Nerd-Folk!
I find it highly suspicious that there's no way to tell from the site where this company is physically located or how to send them non-electronic mail. There isn't even a fax number, just an email address.
I play Nerd-Folk!
Does it need to run a full race? Are they running against each other or just timing laps? Can it be programmed to knock Jacques off ala Schuey? What about if it needs to run in the wet?Is it smart enough to get out of gravel? If so, then maybe it should run in place of Verstappen next year. 8^)
The "F1" appears only in /. article - the site itself calls it "AI-Indy" - which suggests that it's "Indycar" racing. This is much easier than F1 for AI, with a wide oval track being used. Not that it's easy in an absolute sense.
I've never seen a bigger hoax in my life! They haven't even produced anything yet and they're already talking about $$$ & marketing potential.
Anyway, no driver in their right mind would get on the track with a "robotic driver" because the robot is not going to be programmed well enough to perform well *and* also avoid crashes. I wouldn't be surprised if they can get a "robot" to drive around a track (why you would need a human-like robot to do that I don't know... just put cameras on the roof and rewire the steering & acceleration), but there is no way that they're going to be able to get it to drive reasonably (i.e. avoid other cars)
I remember reading a few years back that McLaren tested a remote control F1 car on a race track. There was very little press about it other than some brief mentions in racing-related magazines.
I don't think this robot racer will be too much of a stretch. F1 have lots of on-board systems that can easily be plugged into a central computer. The cars are all fly-by-wire for example.
They used to have Active Suspension, as opposed to Reactive Suspension that exists in a lot of the newer luxury cars. They would actually scan the road ahead and prepare themselves for the various bumps.
The semi-automatic transmissions are another element. The transmissions can actually be programmed to be fully-automatic. Most drivers avoid using this because they feel it takes away from their driving skill. The track parameters are loaded into the transmission system from previous test sessions and the computer takes care of the shifts. Apparently Michael Andretti used this very often when he drove for McLaren. If I'm not mistaken the race he had his best finish was actually a track that the engineers had the most data to load into the various automatic systems.
Another area that has greatly improved thanks to computers is in engine-mapping and traction control. Thanks to the fly-by-wire and advanced engine management systems the drivers used to (not anymore it's illegal now) keep their foot practically planted and the car would modulate the throttle for them. There is word that special programs used to exist just for the standing starts. There was a big rumour going around when Schumacher drove for Benetton that the engineers would always go through a very special routine on a laptop connected to car just before the start. The engineer would double click different areas on the screen and press weird combinations of keys and next thing you know Schumacher had another perfect start.
Formula 1 is a strange series. I don't know if Bernie Ecclestone and the boys have figured out if the Driver's Championship or the Constructer's Championship is more important. I think they realize that the driver's championship offers the short-term attraction to the fans but want to keep F1 at the technological pinnacle. As such they encourage (then ban) highly innovative engineering.
Given all this and the elapsed time that has passed since that era (approximately 5 years ago) I think the AI car will do okay but it will not be post a faster lap than Michael Schumacher. That's besides the fact that the car will most likely have systems that are considered illegal in a Formula 1 car today.
Even Schumacher (who loves to fall back on the "my car is inferior" line) has admitted that, at least at certain times, he has had the better car this year.
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.
ok, maybe the AI isn't totally ready for this type of "race" yet, but how about fitting up one of thse arcade racing machines (you know, the ones you sink your quarters into) and have experiement with having a contest between a real F1 driver and a computer game freak? link up the game to a real car thats active on the grid, add a camera and force feedback controls.. and wola. your in the race?
The web page doesn't say if the robot car must run in race conditions, i.e. in traffic and changing track conditions (oil and rain on track, changing temperatures, changing wind). For unexpected situations, it would be hard to match any human.
I believe it was Juan Manuel Fangio (3-time World Champion in the 50's) that came upon an accident on the far side of a blind corner. However, even though he could not possibly see that the accident had happened, he slowed down before entering the corner and avoided the accident. How did he do it? He noticed the fans in the stands (who could see the accident) were all turned toward the accident, rather than watching his approaching car, thus signalling that something was amiss! Let's see a computer do that!
...can the computerized system deal with a spinout? Or will it just tumble to its oblivion?
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
With costs in formula one already spiraling out of control, this does open up one very interesting possibility - robotic test drivers. Perhaps they could be programmed for the quirks and habits of specific drivers and let loose to run hundreds of test laps a day, though with restrictions on testing in place, it wouldn't help much...
:)
But then again, why not pay the russians to build a test track at Chernobyl... where radiation is so high that no FIA delegate can approach... and test there in unlimited laps...
It's an idea...
Karma makes sense. It makes a lot more sense if you add reincarnation.
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
Here is a snippet from a page at Atlas F1:
Italy also gave Formula One its only two female drivers of the modern era: Maria Teresa Filippi, who made three starts in 1958 and '59, and Maria Grazia ("Lella") Lombardi. Lombardi had a slightly more extensive career, starting 12 times during the 1974-'76 period. She also has the distinction of owning the lowest point count of any of the 279 drivers who have scored at all in Grand Prix racing. This oddity occurred during a problem-plagued 1975 Spanish Grand Prix. Through attrition in front of her, Lombardi had worked her way up to sixth place, and then the race was stopped due to a serious accident. Since more than one-third of the race had been run, half the normal number of points was awarded for each place; Lombardi received one half of a point for being in sixth place at the time.
Molly.
The big red flags for me are the claims that it will be able to beet the F1 champion when it's supposadly finished. If, by some mirical, they were able to get a self-driving car that could make it around the track there's no way they could program for the infinate number of vatiables, like weather and pit stops and other drivers actions. They are almost saying that they could put the equivilent knowledge and experience of an Ayrton Sena in the car.
Deep Blue was able to play chess because there are a finite number of moves that can be made based on the moves previously made. (As a side bar, Kasperov choaked! He should have beaten Deep Blue.)
Maybe when someone invents the positronic brain this will be potentially possible. Untill then this "research" company is just trying to steal money.
---
--
If I actually could spell I'd have spelled it right in the first place.
Deep Thought/Blue is not 'limited to the "brute force" approach', while it does check moves as fast as it can (which any player, human or computer would want to do)
what makes Deep Blue so good is a concept called Singular Extension. What signular extension does is to identify potentialy important moves
and spend time searching in that portion of the game tree, as opposed to 'simply searching as big an area of the tree as possible' as you state.
There is a very good SCIAM article on how DT/B operates.
Singular extension can be increadibly powerful in practice. I had the good fortune to witness a game that DT played in the North
American ACM Computer Chess Championships betwee Deep Thought and Hitech (the other CMU chess computer).
Both team set up their monitors in the same conference room so the spectators could watch the 'thinking' of two programs.
In the end game their came a point where Hitch thought it was up a pawn, but Deep Thought saw mate in 12 moves!
Hitech went on for I think 6 more moves before it decided to resign.
This concept can obviously be applied to the original topic of this discussion, a computer driving a car.
The computer could realize that the car ahead which is begining to spin is a much more important concern
than whether or not the car behind can pass on the next turn if the turn will be blocked by the spinning car
-jon
Are a fasinating bit of technology. You can read all about them in this paper that my father presented at the Fall 1998 AIAA Symposium in Boston.
-jon
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%?
> OK, so sue me. I was a SF geek long before I ever touched a computer.
:) Dont' say "so sue me", even if said sarcasticaly. The US has an extremely high number of lawyers per capita, which a certain segment of the population likes to use.
Tip to American Tourists
Things that make you go hmmm... why are most politicians usually lawyers first ?
Cheers
Mad Scientist thinking: Weeeeeell, time for my next Mindstorms robot. What should I do? I HAVE already done the whole "Does the light stay on if the refrigerator is off thing.....
-----
Score 3? For what? Being wrong, at length? - smirkleton
The Robot watches the light and gets away just that much quicker. The robot watches the rear wheel and controls the wheel spin, same in reverse for braking. It would have anti lock just as a driver would if it could. With the proper input the robot should be quicker. Fuzzy logic anyone?
There isn't enough programming in the world to define a reliable model of what allows a human being to come to harm. I'd suggest going to DragonCon and attending the robot bash: they have a panel that has actual robotics experts/researchers that get into in-depth discussions of what's going on in the field.
There are (broadly speaking) two types of robots: autonomous and non-. Non-autonomous are simply remote-control vehicles. They have no decision-making capability. A fully-autonomous model would be utterly impotent due to bloatware if you tried to implement the Three Laws: remember RoboCop 2? You also have the problem of what harms you might not harm me. That takes a very fine level of discrimination.
SO, IMHO, I don't ever expect to see Asimov's Three Laws implemented in something resembling a viable autonomous robot.
But what I really wish I could see would be Asimov in a panel discussion with actual robotics scientists. THAT would be pretty major.
--
When you sympathize with stupidity, you start thinking like an idiot.
After reading some of the website, I predict that this will never happen, at least not in any meaningful way.
This looks entirely like a publicity stunt. It looks like the commercial aspects are driving it, instead of the technical ones. For example, they appear to plan to have a robot that is in the car. That is purely for marketing reasons - you want an anthropomorphic robot that can capture the imagination of the public and which you can sell dolls of.
In fact, it's crazy to put the computer in the car. Race cars crash. You can't put your Supercomputer at risk like that.
Also, they are outsourcing everything. That is guaranteed to multiply the difficulties. They are planning separate computer hardware, software, and robotics companies. Yeah, right. That's going to work!
And they've already announced a completion date, 2003. I would like this project a whole lot better if it was being done by an established institute (MIT, SRI, PARC) or a supercomputer company (IBM, HP, Compaq) and if it was being done in a low-key, almost invisible way. If this was real, the public wouldn't hear about it until there was a working prototype.
In any event, if I was the world champion F1 driver, I would welcome their efforts to beat my time on a given track. But I wouldn't get on the track at the same time as the robot. No way.
On a similar vein, somebody (I don't know who) said that computers will have as many internal connections as the human brain has synapses in 19 years. I think this comes from extrapolating "Moore's Law" (computer power doubles every 18 months).
I wonder at what stage of complexity self-awareness begins. "I'm afraid I can't do that, Dave."
Look, this is so obviously impossible I can't believe anybody here fell for it. In Japan they are making a really big deal over the fact that they've got a robot that can CLIMB STAIRS. Now you think some Aussies are going to build one that can DRIVE A RACE CAR?!?! COME ON!! The required computer vision technology is decades away. By comparison, landing a 747 is child's play! F1 drivers perform at the absolute limit of human perceptual/physilogical capabilities. There is no way on Earth that any computer vision system could possibly process the data produced by such a situation. Huge gobs of visual data need to be processed on a millisecond by millisecond basis by the brain. We might see computers/robots driving REGULAR cars slowly in 20 years or so but don't count on it. Don't be a sucker--this is the kind of thinking that Jaeon Lanier has been warning people about recently.
Why would anyone willingly be on the same race track as a robotic race car capable of over 200 MPH?
I would prefer to see two robots racing each other for many races before risking a human. At least Kasparov didn't have to worry about a glitch running into him at 200 MPH!
Besides, I would like to see a robot driving a Beetle and trying to beat my wife to a sale;)
yes. this was a bad joke.
Your mind is squeezed by a blast of pain!
These supporters will range from those who make a purchase from our website to others who may wish to make a major personal contribution to the project and take a lap with AI-Andi - at a speed of their choice!
I am not sure I would want to be a passenger in a (two seater) F1 car, with a real F1 driver, let alone a plastic one.
This is a scam/hoax/jinx.
How many billion calculations per second will it take to keep the thing on the road at racing speeds ? more than a beowulf cluster of P4's could muster I would bet !
Not Fragile
I'm not sure how many Formula 1 fans read slashdot, but when I saw this post I was instantly intrigued. I thought I'd add a bit of personal opinion to this. No matter how good this computer may be, Michael Shumacher is probably THE best driver in the world. He has won (I believe) 3 world championships and this past year won even though the second best driver in the world, Mikka Hakinen, was driving a car that was superior to Schumacher's. If anyone could outrace a machine, Schumacher would be the one. Move over Kasparov, we actually have a man who's better than a machine.
Yes, your German sucks (and I'm not a native German speaker), but what's worse is that you and the moderators don't know their Kraftwerk lyrics by heart.
What is the world coming to?
.-.
Help savingAmigaOS and a free PowerPC market
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
Are you really suggesting that having an arm move pieces on a chessboard is hard to do these days?
Ah! I must take exception to this! No, having a robotic arm move pieces, in normal playing conditions, is not difficult. As a tournament chess director, movement of the pieces at high speed is critical in the final phases of a time control ("time pressure"). I think you might have reluctance of certain players, particularly Grand Master Walter Brown (founder of the World Blitz Chess Association) who almost always gets into time pressure (he seems to revel in it!). If I relied on my hands as much as these guys do, I'd be very hesitant to have a robotic arm moving and removing pieces from the board at fast speed: I'd risk injury much more than against a human opponent if our paths crossed: keep in mind that not only must you move your piece, you must remove captured pieces. Brings the touch move into a whole new light!
Invariably human/computer matches, even computer/computer matches, have human "handlers" to move the pieces: combining the chess algorithmic and robotic programming crosses disciplines (plus the crossover between computer programming and mechanical engineering) and would be more than a little bit tricky.
I'll add one more wrinkle: players must write their move with the same hand that they move the piece and operate the clock: so not only does the arm have to move, it has to punch the clock (no big feat, unless the opponent "accidently" shifts its position) but it would also have to write.
There are very specific rules on computers for chess tournaments and the only time you'll see people playing against computers is in exhibition events: they are not official games and are not rated.
Personally, I'd like to see IBM's chess computer in a true championship format: having to play 20 people or more in different time control formats. Their computer was optimized to beat Kasparov: how would it do against Boris Gulko, or Joel Benjamin? (Joel isn't a fair comparison: he was a consultant on the IBM project.)
--
When you sympathize with stupidity, you start thinking like an idiot.
Geez. They really want to cheat, don't they? I can make some static maps on a z80 that can accomplish that. If I know exactly where I am and exactly what the status of my tires are and have a nice Bosch motronic control system with a Ferrari F1 manumatic and some simple shifting pattern maps, then all I need after that is some simple lines to *always* take, depending on the track conditions, and there's no way any human could beat it.
As to handling wet track and skids, my 1996 Camaro can do that...
A society that will trade a little liberty for a little order will lose both and deserve neither. - Thomas Jefferson
The university's robotics group that's affliated to this venture/spam went to the robocup 2000. Their team is called All Boltz. I can't remember their exact place (and the robocup site is down) but I 100% sure that they're one of the teams who score the least.
Hum well as i see it the firstb thing we need to do i make shure schumaker dont cheat cus thats all he is any good at he certainly can't drive he should have been black flagged at least twice in every race so far this season .
use GAs and neural nets. they can learn. yep..its costly but with enough $$$ its possible.
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.
Someone met up with Douglas Adams at some sort of party, and asked why 42 of all the numbers or reasons he could have picked. He replied (very paraphrased)
DA: what is the sum of all the numbers on a normal 6 sided die?
Person: 21
DA: How many dice are thrown in a game of craps?
Person: 2
DA: What is 21 x 2?
Person: 42
DA: There you have it. Life, my friend, is a crap shoot.