Please create an account to participate in the Slashdot moderation system

 



Forgot your password?
typodupeerror
×

Wii Hacked To Control Sword-Wielding Robot 136

ianchaos writes "WiiBot is the pet project of two engineers who apparently have way too much cool hardware and time on their hands. These two guys figure that as long as you have a Kuka KR16 industrial robot to work with, why not see if you can control it with the Wii Remote? The result is a tennis-playing, sword-wielding mechanical arm that simultaneously captures 'weekend of nerdy fun' and 'accident waiting to happen' in a fun two minute video. The website even details the technical aspects of teaching a robot to parry."
This discussion has been archived. No new comments can be posted.

Wii Hacked To Control Sword-Wielding Robot

Comments Filter:
  • The perfect crime? (Score:4, Interesting)

    by Umbral Blot ( 737704 ) on Sunday February 04, 2007 @03:08AM (#17879422) Homepage
    Murder by remote controlled robot, the perfect crime?
  • Neat Implications (Score:5, Interesting)

    by slib ( 876774 ) on Sunday February 04, 2007 @03:25AM (#17879484)
    This could be great for amateur robotics. Instead of painstakingly programming the subtle nuances of motion into a robotic arm/leg/whatever, perform the movement via WiiMote and record the motion. Although I'm sure people have been using similar methods for ages, using a Wii only requires an investment of 250 USD. Plus tax. Plus Zelda (purely for scientific purposes).
  • Next step (Score:2, Interesting)

    by Anonymous Coward on Sunday February 04, 2007 @03:47AM (#17879534)
    I have absolutely no idea what sort of data has to be pushed via serial to the robot, but it seems like the natural next step would be software that translated the tuple of accelerations per axis from the wiimotes accelerometers into the appropriate serial data. This may, of course, be impossible with this robot, and you may only be able to specify a path to follow, after which the robot returns to its original position. However, if you can give the robot commands to move a certain direction relative to its current position, it seems like this would be possible. Depending on the sampling interval from the wiimote, it seems like more natural motion. To address the need/desire to stop the software from listening and the need/desire to have the robot, at some point, return to a standard position, button input could be used (i.e. A starts and stops the software from listening to the wiimote input, B tells the robot to reset to it's default position.

    Again, none of this may be possible with this robot, but that would be a much more interesting hack. It might actually allow you to hit a ball that's already been thrown (and decapitate someone in real time).

    -Lee
  • by I'll Provide The War ( 1045190 ) on Sunday February 04, 2007 @04:49AM (#17879700)
    They should get together with these guys and start charging:

    http://video.google.com/videoplay?docid=-747250219 7006303244 [google.com]
  • by srussia ( 884021 ) on Sunday February 04, 2007 @06:05AM (#17879946)
    I fear the military applications of this...not like it wasn't possible before, but perhaps this might give some people ideas that would ultimately be used to kill people.

    I fear the /.-meme inducing virulence of this comment. Think about it--"I fear the military applications of this..." is the new obligatory "I, for one, welcome our new $ACTION overlords." And it can apply to virtually any slashdot story.

    Example:

    SCIENCE: MATERIAL TOUGHER THAN DIAMOND DEVELOPED

    Obligatory...

    I, for one, fear the military applications of this, not like it wasn't possible before, but perhaps this might give some people ideas that would ultimately be used to kill people.
  • Re:Neat Implications (Score:2, Interesting)

    by bensch128 ( 563853 ) <bensch128@@@yahoo...com> on Sunday February 04, 2007 @06:22AM (#17880000)
    This meant that the robot could do a handful of simple, pre-defined motions, and the Wii-Mote was simply used to select the closest available match.

    They could do better though.

    They should record only very small motion paths (its accurate to 1/100 inch occuring TFA). Then have the robot play each piece when it receives it from the Wii-Remote. They'd have to record alot more motions but it would be smoother, more immediate, and more realistic in the end.

    Cheers
    Ben
  • by mrnick ( 108356 ) on Sunday February 04, 2007 @06:24AM (#17880010) Homepage
    In my graduate studies (computer science Texas A&M U @ Commerce,TX) I have been working on a mini robotic submarine that uses accelerometers to calculate acceleration (of course), speed, and location. We also have a similar arm that was donated to the Physics department but I have not had a chance to play with it yet.

    Anyways, my question is why use wii controllers? You could order a dozen accelerometers from DigiKey for far far less than the price of the wii controller, assuming you can purchase them without buying a complete wii system. Not only would it be cheaper but you could place the accelerometers on your arm, hand, shoulder, etc in locations that most mimic the articulating parts of the actual robotic arm.

    Only thing I can guess is that by using the wii controller, and it being so new, that you get that whole 3lit3 haxor effect because wii and it's use of accelerometers is new for game controllers. Looks like it worked, getting their site slashdotted and all.

    On a side note I did use a wireless xbox (not 360) controller to control the mini robotic submarine. Did you know that those controllers are USB, Microsoft just used a non standard port? It was my intention to have events in the sub send data back that would control the force feedback in the controller but I have not been able to figure out how to send data to the gamepad to activate that function. If anyone has suggestions please let me know (sshscp@gmail.com). I am using the XBCD driver on a windows XP laptop running a program written in G (Labview 8.2.0).

    With this stuff you have to watch what you call things. If something is completely remote controlled then you can't technically call it a robot. At minimum a robot has to be able to act autonomously or perform pre-programed tasks. Currently the sub has a pre-programed task - when battery levels reach a preset point control ballasts and pumps to quickly return to the surface and then go into power consumption mode. It sounds like this arm is a robot since the controller activates a series of pre-programmed movements. But, if it were completely controlled by the accelerometers then it would be a remote controlled arm and not a robot.

    Future plans for the sub are to include autonomous mapping of it's environment. As for the arm I was thinking of working on it so that it could play chess, possibly against live opponents over the web (with a webcam). Do you think people would want to play it?

    Anyways good job on the arm guys! I can't get enough of these micro controller applications. I got the bug quick after having code that I wrote actually do something in the real world for the first time. Maybe I'll get something slashdotted someday ;)

    Nick Powers
  • Re:Neat Implications (Score:2, Interesting)

    by Falladir ( 1026636 ) <kingfalladir@yahoo.com> on Sunday February 04, 2007 @06:26AM (#17880022)
    From the wii games I've played, it seems like the wii isn't capable of anything more than execution of scripted motions either. It's fun to slash by making a slashing motion, but it would be a hell of a lot more fun if the precise motion made by the user appeared in the game. I can't help but feel that if my wii motions are scripted, I might as well have a keypad. This isn't off topic: these guys have come just as far as most of the wii game developers in harvesting data from the wiimote.
  • Re:Looks like... (Score:2, Interesting)

    by jorgevillalobos ( 1044924 ) on Sunday February 04, 2007 @11:14AM (#17881016) Homepage

    Allow me to be the first to tell you that this is one of the coolest things I've seen in a long, long time. It's an amazing idea and one of those things that just make sense :).

    Can't wait for the gun mode :P.

  • by freeweed ( 309734 ) on Sunday February 04, 2007 @11:45AM (#17881216)
    The Wiimote can't give you accurate position data, so thats pretty much all you ever get.

    No, but the Wiimote can give you amazingly accurate ORIENTATION data, which is all you really need for this sort of application. I think the problem here is that these industrial robots aren't designed for real-time control, hence the "mimic" type actions.

    Believe me, the Wiimote can very accurately, in real time, determine exactly which way it it pointing. It's a lot more complicated than "pressing a sword swing button". Just because Zelda was a Gamecube port, and therefore unable to really take advantage of the Wiimote, does not mean the controller is actually that limited in function.

"More software projects have gone awry for lack of calendar time than for all other causes combined." -- Fred Brooks, Jr., _The Mythical Man Month_

Working...