Re:A Robot That Balances on a Ball
by grk on Saturday May 01, 2010 @01:09AM
(#1310148)
You can make your own quite cheaply - with LEGO Mindstorms. I bought a set for my dad last christmas and he's been playing with inverse pendulum ever since. :)
Re:A Robot That Balances on a Ball
by grk on Saturday May 01, 2010 @04:12AM
(#1310286)
The vid I linked to is not my dad's. I think he got a segway-style 2 wheel robot to work. Took him about two months, but he doesn't have much spare time. The biggest problem was adjusting for inaccuracies in the measuring equipment.
Re:A Robot That Balances on a Ball
by ttol on Friday April 30, 2010 @12:17PM
(#1308573)
Prerequisite before usage: Extremely smooth hard surface with close to zero defects.
Warning: Use on soft surface may result in the machine dislodging itself from the ball. Also, any obstructions, such as holes or other obstacles, may result in machine failure.
Re:A Robot That Balances on a Ball
by evlapix on Friday April 30, 2010 @12:34PM
(#1308630)
Typically the 3 omni-directional wheels makes direct contact with the ground. Even then, holes and obstacles are a concern.
I didn't read the article, but judging by the video I see this more as a proof of concept. The balancing doesn't so much impress me. I liked what I saw of them being used together. If you scale the design up by any amount and increase the friction/traction between the omni-directional wheels and the ball (possibly by using a hard rubber surface that has some give to it) you would have a hard time finding a more capable form of omni-directional locomotion.
Re:A Robot That Balances on a Ball
by jdrock on Friday April 30, 2010 @10:25AM
(#1308227)
I wonder if this would make bi-pedal movement obsolete, or at least less important to innovate. There would still be need to move up and down steps, but this form factor seems much more compact than current bi-pedal robots.
Re:A Robot That Balances on a Ball
by abecedarius on Saturday May 01, 2010 @05:19AM
(#1310340)
Make a fancier version that can hop. (When this robot was sketched out in the "Inventions of Daedalus" column in Nature a few decades ago, he went into the how-to-hop issue too, though I don't remember any details.)
Re:A Robot That Balances on a Ball
by foomanchi on Friday April 30, 2010 @10:56AM
(#1308298)
one major drawback to using a rubber coated ball like that is that it will pick up all of the gunk that it rolls over, and those omni wheels have a very low tolerance for obstruction before they get jammed.
http://www.youtube.com/watch?v=f8jxGsg3p0Y this is a lego robot balancing on a ball.
Warning: Use on soft surface may result in the machine dislodging itself from the ball. Also, any obstructions, such as holes or other obstacles, may result in machine failure.
I didn't read the article, but judging by the video I see this more as a proof of concept. The balancing doesn't so much impress me. I liked what I saw of them being used together. If you scale the design up by any amount and increase the friction/traction between the omni-directional wheels and the ball (possibly by using a hard rubber surface that has some give to it) you would have a hard time finding a more capable form of omni-directional locomotion.
I don't think this is meant to be innovating in the "this is going to change the world" sense, but more "look at this cool robot I made!".