Forgot your password?
typodupeerror
Robotics Science

High-Speed Robot Hand Shows Dexterity and Speed 133

Posted by kdawson
from the think-fast dept.
An anonymous reader tips a blog posting that begins "A few blogs are passing around videos of the Ishikawa Komuro Lab's high-speed robot hand performing impressive acts of dexterity and skillful manipulation. However, the video being passed around is slight on details. Meanwhile, their video presentation at ICRA 2009 (which took place in May in Kobe, Japan) has an informative narration and demonstrates additional capabilities. ... [It] shows the manipulator dribbling a ping-pong ball, spinning a pen, throwing a ball, tying knots, grasping a grain of rice with tweezers, and tossing / re-grasping a cellphone!"
This discussion has been archived. No new comments can be posted.

High-Speed Robot Hand Shows Dexterity and Speed

Comments Filter:
  • by Anonymous Coward

    Welcome our new robot overlords.

  • I can high five a robot and get it to do it PROPERLY.
  • holy crap! (Score:1, Interesting)

    by ILuvRamen (1026668)
    Watching that, I can't help but think that the math/programmer behind it is fairly simple with a basic knowledge of physics and the only reason this wasn't around 10 years ago was chips not being fast enough. Now that the technology exists, all the theoretical robot motor function logic can actually be put into practise. If you think about it, the dribbling code must be really simple. Run the numbers while the ball is in the air about precisely where and at what angle the ball impacted the first finger a
    • Re: (Score:2, Insightful)

      by Anonymous Coward

      The programming is (kinda) simple but the math is not. Solving the equations of motion that quickly does require good computers, but also good (aka fast) solvers(algorithms). While these algorithms have been known for about 30-40 years, they haven't been used to their full potential. Of course the robot must have failed the tests initially, but the great thing about science is that once you get it right it stays right (within certain limits).

    • Re:holy crap! (Score:4, Insightful)

      by Cassius Corodes (1084513) on Monday August 24, 2009 @03:26AM (#29170199)
      I wouldn't think so - the amount of adaptability required for the actions would preclude a straight calculation (tiny variations would blow out) - it would more likely be some kind of neural network based approach.
      • Re:holy crap! (Score:5, Insightful)

        by noidentity (188756) on Monday August 24, 2009 @03:57AM (#29170313)

        I wouldn't think so - the amount of adaptability required for the actions would preclude a straight calculation (tiny variations would blow out) - it would more likely be some kind of neural network based approach.

        Feedback. As long as the error for each iteration (bounce) isn't too great, the long-term error can be kept within this by adjusting the next response based on feedback from the previous. Anything that's open-loop (lacking feedback) will fall apart, neural-net-based or not.

        • Re: (Score:3, Interesting)

          by sexconker (1179573)

          No. A closed loop can be perfectly fine forever.

          If the loop is:

          Track ball
          Hit ball when in range

          Then as long as all hits result in the ball being in range, you're good.
          You don't need to measure the error of the last hit - just get the next hit right. Always aim for the most optimal spot for the next hit.

          If the ball ends up further and further away from it's optimal spot with each hit, then the hits are in error (in terms of aiming, timing, speed, etc.).

          You don't need to know what happened last time to be a

    • Re:holy crap! (Score:5, Insightful)

      by wickerprints (1094741) on Monday August 24, 2009 @04:23AM (#29170415)

      Actually, based on the narration, I believe that the computation involved requires three basic processing steps: (1) detection systems to measure physical properties of the system at any given point in time, such as position, velocity, acceleration, and force; (2) real-time algorithms based on rapid numerical solution of equations to predict future states of the system, with continual updating by comparing predicted state with actual state inferred from step 1; and (3) determination of the appropriate movement in the robotic arm for the necessary outcome.

      I think that this is a very difficult thing to program in general because the examples shown are very specific tasks which serve to demonstrate the speed of this type of processing, but we do not see how well arbitrary tasks can be similarly implemented or how accurately.

      Make no mistake: this is very impressive performance, because it is basically a huge step forward in machine vision and real-time robotic control. On some level, the mathematics has always been there, but only in as much as the basic mathematics of binary arithmetic has been used to develop programming languages. There's a lot more going on behind the scenes that extends beyond a mere physical description of the system in question, because for such an approach to be possible in the general sense, the robot doesn't know things like the precise distribution of the mass in the object being manipulated, or all the frictional forces involved. It's not operating under a sort of Laplacian notion wherein if one knew the precise state of all parameters of the system, one can simply solve the required physical equations and predict the future state at any arbitrary point in time, because (a) chaos guarantees the instability of such nonlinear systems, and (b) it wouldn't be possible to measure all such parameters with sufficient precision.

      What is really going on is perhaps best explained in human terms: the programming is doing a lot of what humans do--we observe the state with our visual and tactile senses, and our brains receive these continual updates and decide what to do next. This processing is already extremely fast in a biological context, but with these machines, it is made at least an order of magnitude faster. The next step is to simulate a sort of adaptive intelligence to allow the handling of a wider class of scenarios than the ones shown in the video.

      • by nog_lorp (896553)

        The difficult part is probably the parallel processing needed to achieve these speeds.

      • by bar-agent (698856)

        What is really going on is perhaps best explained in human terms: the programming is doing a lot of what humans do--we observe the state with our visual and tactile senses, and our brains receive these continual updates and decide what to do next. This processing is already extremely fast in a biological context, but with these machines, it is made at least an order of magnitude faster. The next step is to simulate a sort of adaptive intelligence to allow the handling of a wider class of scenarios than the

  • Skynet (Score:5, Funny)

    by vonFinkelstien (687265) on Monday August 24, 2009 @02:17AM (#29169871)
    Seeing just how blazingly fast that thing was makes me know that we have absolutely no chance against Skynet.
    • Re: (Score:2, Funny)

      by jarocho (1617799)
      A robot hand being able to bounce a ping pong ball doesn't make me worry.

      However, if and when a robot learns to play Eddie Van Halen's "Eruption" [youtube.com], then I'll worry.
    • Re: (Score:3, Insightful)

      by Joce640k (829181)

      Yeah, well... they didn't tell us how many times it *dropped* the cellphone. I bet one good catch out of 1000 wouldn't have impressed you quite so much.

      • by necro81 (917438)

        I bet one good catch out of 1000 wouldn't have impressed you quite so much.

        Oh, I don't know - it's still pretty freakin' impressive. One in a thousand ain't bad for something as difficult and improbable as that. Kinda like when I manage to knock a 3-point shot in basketball.

      • by iamhassi (659463)
        "they didn't tell us how many times it *dropped* the cellphone."

        Actually I'd rather it catch an egg and not break it. *That* would be impressive.
    • its like fast zombies.
    • Re: (Score:1, Redundant)

      I propose an addendum to Godwin's Law in which it is stated that as any online discussion about robots grows longer, the probability of comparisons involving Skynet approaches 1.
      • by bar-agent (698856)

        Please. On Slashdot, a Skynet comparison is the third post written, and the first +5 Insightful one!

      • It seems the mods have a great sense of irony -- the guy making the extremely overdone Skynet comment gets modded up (+5 even!) and I get modded down (as redundant of all things!) for pointing this out. WOW!! I can see from the other comments that there are some people who find my thoughts valid -- I call bollocks on whoever modded me down here.
    • Re:Skynet (Score:4, Funny)

      by necro81 (917438) on Monday August 24, 2009 @08:28AM (#29171675) Journal
      Yeah, the neatest trick I've seen a robot pull off since Bishop's knife trick in Aliens.
    • by dissy (172727)

      Seeing just how blazingly fast that thing was makes me know that we have absolutely no chance against Skynet.

      Well, look at the up side. We only have to hide in fear from the machines for one year, and then the series will get canceled by Fox at season two and humanity will be safe once again!

  • Title... (Score:5, Funny)

    by oldhack (1037484) on Monday August 24, 2009 @02:19AM (#29169891)
    That's just so WRONG.
  • by dmomo (256005) on Monday August 24, 2009 @02:19AM (#29169893) Homepage

    For the last decade, I've been eking by on the "well, can a robot tie it's own tie?" Hey, at least there's still "sleeping one's way to the top". And once robots learn to do that... is anyone REALLY going to want to leave their house for a stupid JOB anyway?

  • one grain of rice? (Score:1, Insightful)

    by Anonymous Coward

    I want to see how fast it can move a whole bag of rice. Very impressive, hadn't seen the last few examples before.

    • Re: (Score:3, Funny)

      They should teach that thing KungFu and have Jackie Chan fight it. I just want to see how fast it can rip out a human heart.
  • by rumith (983060)
    My good robotic overlord, why are you tossing that paltry cell phone for three hours in a row already? *Sigh* I guess my inquiries aren't welcome here, sir...
  • The claw (Score:5, Funny)

    by Thanshin (1188877) on Monday August 24, 2009 @02:24AM (#29169913)

    "And inthis sequence this video you can see the robot hand strangling dr. Kamakuro.

    Notice how the pressure sensonrs allows it to know when to release to leave the doctor unconscious but alive.

    Observe the marvelous precision displayed as it cuts the doctor's hand and peels its skin to make itself a costume.

    Ohh, it's trying to sew itself to the doctor's stump; ain't it the cutest thing?"

    • Re: (Score:3, Funny)

      by Thanshin (1188877)

      I sincerely apologize for the eye-bleeding spelling horror I managed to reach with the previous post*.

      Once again, I promise not posting on slashdot before the first dose of caffeine.

      *: My first spelling of that as 'precious post' probably means I need even more caffeine.

      • Observe the marvelous precision displayed as it cuts the doctor's hand and peels its skin to make itself a costume."

        *: My first spelling of that as 'precious post' probably means I need even more caffeine.

        Wasn't "Precious" the name of Buffallo Bill's dog in Silence of the Lambs? Freud would have been elated to have met you...

    • by j-stroy (640921)
      Hook this arm up to the emotive stuff they are doing with face/intent interpretation and expression and you should be able to "tap out" if it squeezes you too tight, unless its a ruthless torture robot.

      I never really connected a purpose with the robot emotional interpretation and expressive stuff before, but I can see a good use for it now. Alternatively, torture robots are self explanatory. Also, I wonder if it can do back massage?
  • Impressive.. (Score:2, Insightful)

    The accuracy is pretty impressive and will definitely get adopted in future robots, though the speed is a bit scary. At least you shouldn't come within its range in the hope that it will follow the three Laws of Robotics.
    • Just do an Indiana Jones on it.

      Mr. Robot stands there hands whizzing around juggling cellphones and manipulating grains of rice at lightening speed, then you pull out a shotgun and pump his CPU full of bird shot.

    • by rrohbeck (944847)

      The accuracy is pretty impressive and will definitely get adopted in future robots, though the speed is a bit scary. At least you shouldn't come within its range in the hope that it will follow the three Laws of Robotics.

      Many modern robots work at that speed. I worked with tape handling robots for years and the ones that are large enough that you can stick a hand (or your whole body) into them have safety switches that disable the robotics or slow them down so that you can get out of the way and momentum is eliminated.
      Of course, engineers disable those switches. I've been hit on a hand hard more than once, and I know of one guy who was hit on the head by a tape robot. He went to the hospital but didn't need stitches.

  • Dribbling demo? (Score:3, Interesting)

    by QuoteMstr (55051) <dan.colascione@gmail.com> on Monday August 24, 2009 @02:53AM (#29170023)

    Very impressive. I wasn't able to quite tell from the video, though: was the end of the dribbling demo planned, or did the robot lose control of the ball after a few seconds?

    • Re: (Score:1, Informative)

      by Anonymous Coward

      dude the dribbling lasted less than 100ms....the ball is bound to eventually bounce away because of even slight error in the visual recognition, which was quite cool.

      Imagine an evil robot that could do something that fast, it would be able to dodge bullets.

    • I was wondering the same thing. I came to the (completely uninformed) conclusion that it was not planned.
      • About your signature:

        If you can read this... 01110101 01110010 00100000 01100001 00100000 01100111 01100101 01100101 01101011

        I guess you didn't know how to enceode 'y', 'o' and '\'', right? :-)

    • by xigxag (167441)

      Playing it back frame by frame, it looks like the ball made a weird bounce and then skittered out of range of the robot, which sbusequently threw its hands, er, fingers up in resignation.

    • It didn't say if it lost control or was planned. But the dribbling period was only 100ms. But given how fast it dribbles, it would be like some kid dribbling for like 10 seconds before losing it. well, that's just a guess
    • by bar-agent (698856)

      was the end of the dribbling demo planned, or did the robot lose control of the ball after a few seconds?

      Actually, seeing how much the hand's mount was bucking, I was surprised it was able to dribble at all.

      (Reading it back, that sentence sounds vaguely but horribly disturbing.)

  • by Animats (122034) on Monday August 24, 2009 @03:18AM (#29170153) Homepage

    This is very nice work. The most interesting result is that some manipulation problems become easier if done fast. In the short term, inertia makes the motions of objects very predictable. With millisecond reaction times, that can be exploited.

    Fast machinery isn't unusual, but it's rarely that smart.

    • Re: (Score:1, Funny)

      by Anonymous Coward

      This is very nice work. The most interesting result is that some manipulation problems become easier if done fast. In the short term, inertia makes the motions of objects very predictable. With millisecond reaction times, that can be exploited.

      Yeah, that long term inertia is a real PITA.

    • Um, it would be more impressive if the machinery was slow. I mean, it would then require more intelligence to do the same job.

  • So they built a right-handed robot? I hope they'll follow-up with a left-handed one soon, in order to not be discriminating!

    SCNR

    • A right-handed robot is Dexterous. A left handed robot is Sinister. I really hope they don't go making any left-handed robots or we're doomed.
  • new experiment (Score:2, Interesting)

    by StripedCow (776465)

    i have an idea for a new experiment: fire a bullet at it, and see if it can catch it :-)
    it would be the ultimate body guard :-)

    • I've seen a martial arts demonstration where a guy caught arrow being shot past him. He did it **blindfold**, based on hearing the release and knowing the distance! Gotta wonder how many times he got an arrow stuck in his hand before he got the timing right!

  • Thus far, robots have tended to shuffle along in an awkward, spastic manner - and at low speed. Could this kind of development help them out? If the reason for the difficulties with bipedal motion in robots is that the actuators cannot respond quickly/accurately enough to maintain balance well, then it will be able to. I'd like to know how fast such a robot can move, as it can already tie me up and perform surgery on me if it catches me...

    • by rrohbeck (944847)

      It's also a matter of actuator power (and energy storage of course). But you're right, dynamic equilibrium and control is often easier than static. Try to walk in extreme slow motion and you'll know the difference.

  • anyone know what will happen if this technology is combined with http://spectrum.ieee.org/blog/robotics/robotics-software/automaton/robots-evolve-to-exploit-inadvertent-cues [ieee.org] and used for tasks such as food harvesting also, robo wars of the 2 technologies combined would be fun -- superfast and evolving bots...
  • I don't know why this is news, they've had extremely fast, highly dexterous guitar playing robots since the 1980s in the form of Yngwie Malmsteen and Micheal Angelo Batio.
  • Finally robots will be able to do the high speed, high precision moves magicians need to "create" coins out of thin air or do card tricks.
  • T-1000

    I for one welcome the new self aware machine overlords!

  • I am now going to buy shares in the companies that make this, the flesh light and real doll. These three together will make a HUGE amount of money.

  • Funny, I just watched 'The Forbin Project' last night. If you've got an hour or so to kill and like evil computer stories and don't mind watching an old movie made in 1970 with the USSR as adversary you'll like this movie. It was quite good.

    http://www.youtube.com/watch?v=kQjebwUrhvc [youtube.com]
  • I can play the holophoner.

  • A lot of highly-dexterous manual labor has been moved out of developed countries to developing ones because of the high level of developing country wages.

    Where possible, developed country manufacturing companies have eradicated manual operations with machines, to the extent that despite manufacturing output rising, manufacturing employment is falling in the developed world. But some things (such as shirt sewing and shoe assembly) are still impossible to automate currently.

    Should developed countries become

Things are not as simple as they seems at first. - Edward Thorp

Working...