Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Science

AI Monkey Robot 93

Posted by emmett from the curious-george dept.
Sircus writes "The BBC is carrying this article regarding a robot 'monkey' that has been taught to swing from rung to rung of a ladder, using only the distance between the rungs and some basic equations about swinging. There's not much technical detail, but this is an interesting approach to the problem of dumb robots, and one which certainly gives the suits something pretty to look at."
This discussion has been archived. No new comments can be posted.

AI Monkey Robot More

AI Monkey Robot

Comments Filter:
  • I'm sick of this pathetic attempt for escaping the label of monkey by these so called `Apes'. It's clearly a case of twisting definitions around in order to seperate themselves from the general public's typical negative assumptions about monkeys. Apes, monkeys, who cares. Next thing you know people will want to disassociate themselves with script kiddies by calling them crackers instead of hackers. Get over it you ape loving monkey.

    (IFL)

    Bad Mojo
  • ... can we punch those monkeys????
    --
    " It's a ligne Maginot [maginot.org]-in-the-sky "

  • IMO - Not that great... (Score:3)

    by Diamond Slicer ( 39462 ) on Wednesday February 02, 2000 @10:04AM (#1310849) Journal
    IMO, this is not that great. Heres why:

    The article states:

    "The human instructors have told it some equations for swinging and the distance between the rungs - it must do the rest." - The monkey knows how far it has to go to get to the next rung. What would be amazing is if it could sense where the rung is and calculate that distance itself. It does not solve the problem (to quote the article again) of "bumping into the object" because the monkey knows the object is there.

    While it may be great for the mechanics of a robot, (able to swing) it is not that new of a technological developement. I have toured several car making plants. At one plant they had a robot that cleaned the floors and all it had was a map of the building... it simply sensed when objects (people and trash cans) where in the way. The monkey has everything told to it and it also uses trial and error.

    What may be new is the fact that they managed to fit all that info and technology into something fairly small (judging from the pictures). Robots that are small and can do what the monkey does are rare (I do not know of any others that exist). I just don't think the swinging (kick legs/hands out until object is caught - then repeat) is a breakthrough.

    Remember - this is just my opinion (not flamebait or troll) and I may be wrong... but I have seen far better robots (judging by usefulness, size, and technology) around.
  • One thousand monkeys. One thousand MicroVAXes. One thousand years.
    X windows.

    So what if they're real or AI?

    --Ben

  • It's just a 5 second clip showing the robot in motion, but it really is breathtaking!
  • Well, a smaller robot on litle rails (like the library ladders) would be more effective than one on two legs... Two legs is, in general, not that much of an advantage for most activities, and creates far more problems than it solves. Far more balance is required, and given the length/power ratios of our limbs, we are amazingly not efficient. Look at a pet cat - a leap from a floor to the top of the counter, window sill, or fridge. Now think about it. You should be able to leap to the top of a two story house with that same little effort. Or just the following:
    Top Speeds: Man(10-20mph); Cheetah(60-70mph)

    Two legs are a problem for most practical designs, not a solution.

  • You have to remember this is not as simple as reach out and grab the next rung. The robot starts hanging directly beneath a rung and it has to swing to get up enough momentum to get to the next rung. It does this by swinging the legs and twisting at the hips. It had to learn all of this from scratch.

    I think you will find that if they had not told the robot the distance between the rung then it would have to learn it using some kind of visual recognition system which is an entirely different project.

    I wonder how many people here who claim that this is nothing much could do it...

  • Mechanically speaking, it's probably been possible to do something like this for quite a while. What makes this new is that the robot is now able to figure some things out for itself in case it misses. It uses stereo vision to locate colored balls on its joints, giving it a sort of kinesthetic sense. So if it misses a rung, it can adjust and take another swing at it.

    I wonder why they told it the distance between the rungs, though. If the robot is fast enough to track its own movements in three dimensions, what stops it from locating the next rung the same way?

  • Check out this [newbreedsoftware.com] site for proof.

    It's clear that "robo-monkeys" have been researched before, and their usefullness in conquering "evil forces" is at leat partially explored.

    BoboBot was merely a simulation and technology demo; this latest BBC article proves that governments are working to create legions of these littls beasts.

  • I almost posted this story when it was being shown on BBC1 last night when it went out the presenter said "..and we know that Gibbons are apes" she also said the scientists working on the project kept a furry suit in the corner of the artifical gibbon room, but refused to put it on for the cameras.

    Now Philipa Forester without her clothes would be news.

    Sparkes

    *** www.linuxuk.co.uk relaunches 1 Mar 2000 ***
  • No, I dont think it is truly learning. That is the problem that I see with everything people try to put off as "AI" right now. From what I can tell, none of it is intelligence, it's really pattern recognition mapped onto some basic function.
    If it were really the robot having some sort of cognitive action in here, the only things going on in the robot's circutry would be some basic things it likes, some basic things it doesn't like, and the ability to somehow trace what it likes and doesnt like to its actions and happenings in its environment, most likely through chronological proximity. It wouldn't be getting to the next rung because it knows these equations, it would be getting to the rung purely on estimation by realizing "if i reach this far for a rung this far away, I fall. Falling hurts." from there it would go in two directions - either stop trying to get across or keep going, depending on whether or not it has an incentive to get across. There shouldn't be any physics equations or ability to tell exact distances unless the robot supplies them itself.
  • Hmmm... well, according to some random article in one of these mags I have around here, the cheetah will hunt in a 25 or larger mile radius around its home - note that 'home' can move during extended hunts. I suppose if you put a cheetah on an airplane, it would have very good range, indeed 8^) The point of a bike sort of proves my first argument that many animals are more *naturally* efficient than us physically. We still have a small edge in the thinking area, tho...
  • Exactly - we just aren't put together for impressive displays of physical prowess. Horses, dogs, cats (and many others) can run at high speeds through rough terrain due to the extra legs, leg alignment, and the fact that they don't need these huge, floppy feet that get caught on everything... Great for extra balance (and a little bit of extra thrust), but again, quite a hinderance.

  • More info... (Score:4)

    by Space Cow ( 93479 ) on Wednesday February 02, 2000 @10:11AM (#1310864)
    Ok, took me a while to dig this out, but here is a bunch of info from the booklet I received when I visited Prof Fukuda's lab. There are some nice graphs and such that unfortunately I can't include. Enjoy.

    From Reseach Activities 1998, Robotics and Mechatronics Laboratory of Micro System Control, Nagoya University.

    Brachiation Robot That Learns By Doing
    • Mobile robot moving from branch to branch dynamically like a long-armed ape.
    • Motion improvement by off-line learning.
    • Using real-time tracking system as a visual sensor
    • Using kinetic energy of center of gravity as evaluated value

    Study on the Control of a Two-link Brachiating Robbot via "Target Dynamics"
    • Wish to understand how dynamically dexterous taks can be achieved using physical insight into the designated task and intrinsic dynamics of the system.
    • Task is encoded as an output of a target dynamical system motivated by the pendulous motion of an ape's brachiation.
    • Extension to the "Irregular Ladder Problem" - brachiation on a ladder with irregular intervals.
    • Validation of the proposed strategy by numerical simulation and experimental implementation.

  • Compare and Contrast... (Score:5)

    by rafial ( 4671 ) on Wednesday February 02, 2000 @10:11AM (#1310865) Homepage

    ...this approach, with the one taken by Mark Tilden:

    The latest Smithosonian had an especially good article about him. Apparently his original demo 'bot was a robust walker that could clear most obstacles, run by all of 12 transistors.

    I think somebody needs to look at combining the approaches of AI based robots that make decisions based on modeling their world, and robots such as Tilden's that seem to simply have "body feel" for their environment. Perhaps a Tilden type robot body could carry a reasoning "head" that sets goals for the body to carry out. I think this would be much like ourselves. Imagine if you had to walk or jump with only your eyes to guide you. No sense of balance, or kinesthetic sense. I doubt you could do it. We shouldn't expect robots to do the same.

  • from the article: "Balls help it find its way around in space."

    I remember hearing about studies that suggested men are better at visualizing spatial relationships than women. It seems these researchers have finally determined *why* men are better at it: it's all in the balls! :)
  • Wow! Just imagine a barrel of these things.
  • Is that A.I. (artificial intelligence) or "Al" (the Vice-President's first name)? I know "Al" and "robot" fit together well, but what does "monkey" have to do with it all? :)

    Ryan
  • I belive I saw that the reason a cat can jump so well is the way their back leg muscles are designed. they ahve the ability to use them more like springs, something we can't do. At least not till I get my implants put in.
  • The main benefit of a robot with two legs is to eventually make an android out of the thing. Humans might not have the optimal shape for a lot of things, but we still cope. Human-modeled robots will be able to function in an environment designed for humans. A smart car can drive Miss Daisy around, but it won't be able to help her up the house steps. But a smart humanoid robot would be able to drive a dumb car, help her up the steps, and reach the cat food on the top shelf in the kitchen. Having an android means that you don't need three separate robots each optimized for only one task.

    There's a comparison to be made WRT software design- many small tools vs. monolithic apps.

  • So if you take a 1000 AI monkeys and line them up at typewriters will the eventually write hamlet? And, hey with distributed processing we could recreate the greatest works of literature in no time. So who wants to write AI-Monkey@home? :)


    This message was not written by a 1000 monkeys...yet!
  • "pray for mojo"



    ------------------------------------------------ ----------
  • Once again while America was napping our foreign competitors have seized the lead in exciting new technologies.

    The Japanese, who already lead the world in robotic dog technology, are now prepared to lock-up yet another species of mammillia in a flood of patents and trade secrets.

    Pretty soon soon the market will be flooded with cheaply made imported robotic primates. How will the domestic robotic-monkey manufacturer be able to compete with the overwhelming tide of inferior, but inexpensive, cyber-chimps, oragutans and lemurs?

    This is all the result of ignoring my repeated warnings re:NAFTA, the yakuza, the Church of the SubGenius and certain elements of freemasonry.

    JFK died to preserve our rights to quality digital simians at affordable prices made right here in the US of A! And how do we repay him? We're so self-absorbed we don't notice our precious robotic-animal heritage being stolen from us. It took a British orginization, the BBC, to call it to our attention.

    If we don't do anything about this, soon the international mega-cartels will control our mechanical-mammal destiny, and we can't afford to let that happen.

    That is why I am announcing the creation of the open-sourced CyberWombat, which I will be releasing under the GPL. Since I know nothing about robotics, programming (or Wombats for that matter), I will need a little help. Any takers?

    HipNerd
  • I just saw the show. It's fascinating.

    They have been trying to get things to walk, just by letting "trial and error" evolve. This is an advancement.

    Now they are a bit further by having a 14-motor ape swing from rung to rung on a ladder. Some people may be afraid for AIs, but then they should just protest. I mean, this is just another step on the big ladder.

    Roger.
  • I saw this robot on Scientific American Frontiers at least a year ago, so it's not that new.

    Also, swinging along parallel ropes of a known distance appart is nowhere near as hard as walking. If it could literally swing between oddly arranged branches, that'd be cool. But it's not. It's another interesting case study in "how animals move", but the lessons you can learn from this do not translate very easily into the world of walking robots.

    They're fun, because I could see the power company making robots that walk along power lines, or some crazy thing. Imagine a ski lift that swings up the lines. Lots of weird, fun applications come to mind.

    But, this robot is not a "big swing forward" in robot development. It's a very specific, not very useful instance of people with too much time trying to make a robot that does Yet Another Goofy Thing.

  • If an infinite number of swinging monkey robots, along with an infinite number of swinging input devices, translating contact to text, how long would it take them to come out with SP10 for win2k?

  • Now we need to teach it to bang a pair of cymbals together and squeak :)

    But seriously, having one of these around the house would be awesome!

  • What the hell is so special about this I could probably do this with a couple of set of lego mindstorms. The test environment isnt chaotic enough to consider the monkey navigation anything special. The same thing could be done with wieghts and pulleys given enogh time.? I just dont get the point......
  • If only a few simple equations (known for centuries) and the distance between the bars were needed, then why is this such new? Have the technical abilities of robots to time their catches not been up to this before?
  • Robot Monkey? I can't wait to see what The Onion [theonion.com] does for this one; what I'm referring to is the take on the Aibo [slashdot.org] that they did several months ago.

    And if you watch Space Ghost on comedy central, remember, don't let this monkey take notes for you or you'll pass in book reports that read:

    "Hello, my name is Bingo. I like to climb on things. Can I have a bananna? Eeep eep."

    You'll get an F.

    --

  • The trouble with monkeys (Score:3)

    by Bad Mojo ( 12210 ) on Wednesday February 02, 2000 @09:35AM (#1310892)
    While I feel that AI is a noble cause, I question the reasons behind an artifical monkey. What's next? Robot monkeys in congress? Robot monkeys for president? Is no one considering the effect this might have on our children? I'm willing to support the advance of technology, but I draw the line at robotic monkeys. If you recall the US attempted to put a monkey into space and frankly, we all know how that ended ... in utter chaos.

    Perhaps Jon Katz could write an article on how this will ruin life as we know it unless stopped. Or maybe how I'm oppressing robotic monkeys and forcing them to commit violence in labratories across the nation. I can hardly wait.


    Bad Mojo
  • Hmm... Around... say ten minutes? all they'd need to do was change the #declare GPF_TIME_OUT = 10; to something else...
  • ...look like Curious George. I mean, if you're going to have a robot monkey swinging around, shouldn't it look like Curious George?

    - tokengeekgrrl

  • Probably has great applications for walking robots (Score:5)

    by JudgePagLIVR ( 145069 ) on Wednesday February 02, 2000 @09:43AM (#1310897)

    The same type of calculation of forces is necessary for the function of a walking robot. Now, with walking, you're dealing with a push instead of a pull, and feet that have to balance instead of arms that have to grip, but the calculations are similar.

    But most signifigant is that the robot has to try to grab the next rung, and if it doesn't, it has to figure out what adjustments to make on the fly. Simlarily, if a robotic step went wrong, the robot would have to recognise it, and correct itself before it fell

  • The monkey should be taught how to use a typewriter and then mass-produced.

    --

  • Ahhh, but the fact remains that the swinging monkey robot has the luxury of several attempts to grab at the rung until its swinging slows (from what I read it can also accelerate its swinging with its legs). In the case of a walking robot one missed step and it has fallen, therefore the walking robot cannot as easily learn from its mistakes without having some support other than the two legs that it uses to walk.
  • I can imagine this is creating quite a stir among the pointy heads. I mean, these things can perform all of their duties and at a much reduced cost!
  • It can never truly imitate a monkey unless it throws artificial feces at everyone.
  • What's next? Robot monkeys in congress? Robot monkeys for president?

    Dude, we're already there!
  • Fifteen years does seem like a long time to come up with a swinging robot monkey. However, what is special is that it learns. Sure, I could build a monkey that swings from tree to tree too. But this monkey does so by trial and error, learning how. That seems to be the goal in AI these days. There is just no way we could give any type of robot (at least one we were trying to make intelligent) all the knowledge it could ever need. Programming the robot so that it will learn is critical.

  • But is it truely learning? (Score:3)

    by tepp ( 131345 ) on Wednesday February 02, 2000 @10:50AM (#1310905)
    Ok. So the monkey enters a function to start a swing, and sees how long before it can get it's second hand onto the rung.

    But is it truely learning?

    Or is it just a simple function? Perhaps all it is, besides a heck of alot of expensive equipment, is the following:

    int try_swing(int force, int lastForce, int secondLastForce)
    {
    if (abs(force) > FORCEMAX) return FALSE; // out of reach
    if (force == secondLastForce) return FALSE; // We're just swinging between too high and too low and can't ever make the branch;
    switch (swing_out(force)) // Try to grab the branch with this amount of force in his "kick" backwards
    {
    case GOT_IT : return TRUE; // caught branch
    case TOO_HIGH : return try_swing(--force, force, lastForce); // Try a less force
    case TOO_LOW : return try_swing(++force,force, lastForce); // Try a more force
    default: return FALSE; // Something happened, we can't do it
    }
    }

    My point is not to insult the researchers who created the monkey, but instead to point out that just because it can "adapt" enough to swing from branch to branch, that does not mean it thinks! If that is the case, then the thermostat in your house is a thinking robot, because it turns on the heat when it's cold in your house, and turns the heat off when the temperature is too warm.

    For this monkey to truely "think", he would need to remember everything he has "learned" in the past and apply it to each new branch, or even to other aspects of its life. For instance, if it ran into a tree, the result should hurt it so much that it would resolve to shy away from trees in the future, rather than run into them again. He should recognize trees that he has already seen before and run them flawlessly, and use skills he learned from one tree on the next.

    Notice, I said 'skills'. Anything can be programmed to do one skill. But for it to take that skill and turn it into a new skill, such as taking an unintentional fall and catching the next rung below him, and thus learning how to go from level to level... or to "learn", without ever being programmed, how to catch a branch behind it with its feet...that is learning.

  • What's next? Robot monkeys in congress? Robot monkeys for president?

    Dude, we're already there!

    Sorry to burst your bubble, but the monkeys in washington are entirely organic. Except for Strom Thurmond, he's being held together with duct tape.

  • I don't think it was 'advanced circuit design.' More like feeding back a bunch of AND gates with each other. The fact that it works is pretty cool though.

  • what are the odds that this AI monkey would evolve into an AI human? And if it did would they ban it in Kansas?
    ----------------------------------------- ---
  • There already are walking robots. The Honda P3 [honda.co.jp], probably the most advanced robot in the world. It's not exactly agile, but it does walk on 2 legs without exterior support, and it can even walk up and down stairs.
  • But the problem with lego monkeys is that they fall apart when then hit the next rung... Does anyone else have that problem with mindstorms?
  • Why do I keep getting the urge to punch it?
  • > What exactly is the benefit of a walking robot
    > with two legs?

    The reason you only have two legs: energy is expensive. At least when you are running of batteries or pizza.

    Ryan Salsbury
  • But what kind of range does a cheetah have? (I honestly don't know). I can walk 20 miles in a day (not day + night). Given a bike I can go 100. Maybe humans are more efficient than cats. On the other hand, cheetahs can probably climb trees better than me.

    Ryan
  • No, just fit little wheels to the books, and they will come to you ;-)
  • Ok, you know that we evolved from apes, right? And most apes swing through trees, right? Well, if you think about it, this may be the first step on the way to robots that walk on two legs.

    However, this is in no way saying that C-3P0 is right around the corner.
  • This robot is emulating an ape (such as a gibbon) and not a monkey.

    Monkeys generally walk *on* the branches and leap from branch to branch. (Sometimes hanging from the branches to grab things, etc.)

    Apes are the critters that brachiate. Brachiation allows the animal moving around in the trees to be larger than if it leapt from branch to branch. (Of course, some of the "greater apes" (orangutan, gorilla, chimpanzee and us...) got big enough that they rarely go up in trees anymore -- a full grown silverback male gorilla would break most trees if it tried to climb them)

    I guess "robo-ape" doesn't sound as good as "robo-monkey" -- especially since most people think of gorillas and not gibbons when they hear "ape".
  • C3P0 and R2D2's escape pod land on desert planet, where they discover the grime reality of Robotic Apes that have evolved from R2 Units!
  • Archie McFee [mcphee.com] has had these for years, and a lot cheaper, too. Great for Science Fairs and Department meetings.
  • Hehe, look at how many SP's they had before they decided to come out with WinNT 4.0 from 3.51 :)
  • Cripes! This thing is sure to register as a false positive for those running the YETI@home client. As if sock monkeys and robin williams weren't enough....
  • The quality of TV programs is deteriorating fast, all due to an eeevil plot by Mojo Jojo.

    Meanwhile, the Powerpuff Girls, in colaboration with the BBC is putting out this false article to deceive Mojo Jojo into coming out of hiding so that Buttercup can beat him up again.

    Alternatively, this could be some really intriguing research into real-world-ish dynamic systems, as most Computer Scientists can't handle anything more complicated than stationary cubes.

  • What exactly is the benefit of a walking robot with two legs?
    Could probably be used where there is limited space. Perhaps could be used to find books in a library? (just a suggestion)

    Mikael Jacobson
  • BBC was prepared for nit-pickers like you.
    At the end of the page, they say: And yes we do know that gibbons are apes and not monkeys.
  • Hanging requires no active control, but standing does. While I'm sure there are similarities between swinging and walking, the fact that walking puts the robot in an unstable situation complicates things. (The inverted pendulum requires active control to keep it up, but a pendulum requires no control at all. :)
    ---------------------------------------------- ---------
  • Linus Torvalds is to Linux as Mark Tilden is to robotics.

    For Mark Tilden's Home page: http://nis-www.lanl.gov/robot/

    For the B.E.A.M. Robotics page: http://www.beam-online.com/

  • I guess the next step is to program it with the Dewey Decimal System and teach it how to climb shelves.

  • Analog AI (Score:3)

    by D3 ( 31029 ) <daviddhenningNO@SPAMgmail.com> on Wednesday February 02, 2000 @11:38AM (#1310932) Journal
    Smithsonian magazine (not sure of any link to this) has an article about a guy at Los Alamos named Mark Tilden. He uses advanced circuit design to build a machine that in some ways is much more capable to walk in a random environment than most computer controlled machines. His first 'bot had 4 legs and a total of 12 transistors that allowed it to figure out some complex behavior. Really cool stuff.

  • I want to get a AI monkey and that dam iBO dog robot thing. You know the one, Sony makes it, looks like a dog, everyone and their mom is pimping these dogs out on their web site in contests.

    Anyways, get these 2 robot creatures and put them in a free for all death match live though web cam on the Internet. Progess in the AI robtics field is only good if other AI robots can take them on in a death match, a fight to the finish.

    Plus I am just getting so sick of getting spam for signing up for all these contests to win a dam little robotic dog that doesn't even come standard with a flame thrower. A buzz saw would not be hard to retro-fit on it though...

  • Really cool to see! (Score:5)

    by Space Cow ( 93479 ) on Wednesday February 02, 2000 @09:52AM (#1310934)
    I will be studying under Prof Fukuda starting next fall. During the winter break, I went to Japan and was invited to his lab to see the various projects there. Besides the monkey robot, the projects that were most interesting to me were a cooperative micro robot project, a set of legs that learn to walk, and a feedback control system for an invasive surgery tool/camera.

    The monkey robot is currently most limited by the power/data cables that give it life and can only travel one or two bars before needing to be placed back where it started. It's motions are incredibly lifelike.

    For more information about the people and projects at Prof Fukuda's lab, check out this link to the English version [nagoya-u.ac.jp] of the web page.
  • Then there is also The Great AIP [sourceforge.net] (Artificial Intelligence Project) which is an internet open source project that has started recently.
  • Who's the monkey is now....
  • Does this make Erwin(Re: Userfriendly) a monkeys uncle?

  • Monkey mirror here (Score:1)

    by Anonymous Coward
    See the monkey [dorsai.org].

    Thank you.

Slashdot Top Deals

BLISS is ignorance.

Close