Omni-threads/balls

[timytamy]

I just found this (http://www.buildlounge.com/2011/09/30/omnicrawler-tank-treads-meet-omni-balls/) and this (http://www-hh.mech.eng.osaka-u.ac.jp/robotics/Omni-Ball_e.html).

The first is an omni directional tank tread system, it's driven using omniballs from the second. How well would these two designs scale up?

[Jared Russell]

A ball drive at FRC scales has indeed been done before:
http://www.firstroboticscanada.org/main/wp-content/gallery/robot-mechanisms/336_large.jpg

The omni-tread idea has not been attempted in FRC (as far as I am aware). There would be many challenges to overcome in manufacturing these types of treads. There are also other types of more traditional drivetrains which offer omnidirectional movement that would be far easier to construct. Still, the concept does indeed work at large scales: http://www.youtube.com/watch?v=gpcv_-ICcIk

[JamesCH95]

I bet the "omni-ball" design would scale up well, but they would be much larger and heavier than just using omni-wheels. Omni wheels do the same thing, i.e. drive in one direction while allowing movement in an orthogonal direction.

[Tom I]

For the majority of teams like my own that don't have the capability to mold their own plastics (Oh how I envy Team 357, Royal Assault!), I'm thinking a lacrosse ball cut in half could potentially be a way of cheaply making an omni-ball, even just to experiment.