Slashdot Mirror
Robots With Square Wheels?
Roland Piquepaille writes "About eighteen months ago, I told you about a tricycle with square wheels which needed a specially designed road. But now, Distributed Robotics, a company from Troy, N.Y., is developing robots with square wheels which don't need specific roads. These new 'cars' propel themselves on flat surfaces by taking advantage of gravity. This might sound crazy, but the inventors think it could lead to new robots and toys, and more generally to new micro-machines or MEMS applications."
Because it's a new way to do an old task. Is it practical? I don't know, how many "impractical" or "silly" things later were found to be extremely useful?
Looking at the article I was just struck with a sense of "whoa neat" at the simplicity of the idea yet the fact (as far as I know) it's never been done before (using a "helicopter" of shifting weight to propel a car forward by it's properly aligned square wheels). Sometimes it's not so much "why?" but "why hasn't anyone ever thought of this before?".
--- www.f-theocean.com
I RTA, and I I think this is very neat (if not elsewhere, this could at least be very useful in wheel-and-circle-deficient Lego builds! ;o) ).
Still, I don't see it. Surely, if you keep the axle linkages, sand the wheels down, and merely rotate the motor 90 degrees so its drive shaft is parallel to the axles, that would also yield propulsion without right-angle gearing, no? Granted, you would need a chain drive or a (non-angled) set of gears to connect motor and axle, unless the axle IS the drive shaft.
"Good news, everyone!"
A few moments playing it out in your head will make it obvious that square wheels are actually worse at climbing stairs than round ones. Sitting on them, sure. Climbing them, not so much. A star-shaped wheel might be better, but it'd have to be big (each flat side as long as the step itself) which means a big vehicle, which means a big, heavy swinging hammer jerking the vehicle in every direction on it's way up.
And that still wouldn't work, even if you shifted the hammer forward to keep it from throwing itself down the stairs. The things propulsion comes from each wheel being offset from each other, but being offset for propulsion means they can't all maintain optimal contact with the steps.
This thing is limited to running on a flat, smooth, hard, obstacle-free, level surface, and doesn't even move smoothly on that. Plus swinging a large hammer seems much less efficient than driving a small wheel. It's a curiosity with no practical application. The tag at the bottom of slashdots page ATM is apt: Too clever is dumb.
I just saw this great report on television the other day. It turns out no one wants a Charlie in the Box, a squirtgun that squirts jelly, or a robot with square wheels. They're all just Misfits.
Oh, and Bumbles bounce.
Because when you're making machines at microscopic scales, you get a whole new set of problems. Lubricating bearings is difficult, because conventional lubricants are too viscous. Assembling complex devices is difficult, because you need complex devices to do it. And reliably creating smooth round surfaces is difficult because irregularities in the material cause rough surfaces. Flat surfaces are easy to make - just shear a crystaline material.
If I seem short sighted, it is because I stand on the shoulders of midgets
How exactly can you make a motorized rotating turret without doing all of the things you just suggested were difficult.
Read my other posts on this subject to see that I am already aware of these kinds of problems.
liqbase
Didn't you just define a tank tread? Seriously, ;) isn't a tank tread a closed loop road way that the wheels of the tank drive on?
I am protected!
You may download the video from
e %20Wheel%201.mpg
ftp://mirror.cefetpr.br/pub/misc/Reinventing%20th