A wheel in a ball
The idea we had allows us to have a sphere roll in any direction of the horizontal plane (i.e. rotate the sphere around any vector of the horizontal plane).
The main idea is having a wheel inside the sphere (and in sufficient contact with it) that rolls around any horizontal axis – it would be rotated (slowly enough)around the vertical axis.
Now to ensure the verticality of this axis (and indeed have a whole internal structure that’s only translating and not rotating in the floor’s referential) we thought of having a plane at half the sphere with a ball bearing ring around it. We can ensure horizontality by using a horizontal gyroscope, that hinders any efforts to change the direction of its vertical rotation vector, much like a spinning-top.
We will need to test turning and braking as we don’t want to lose the wheel/sphere contact, so it’ll be like a bike – you can’t turn too sharp or brake to hard, otherwise you’ll get a crash course on the fly on how to fly. The whole thing won’t be too reactive it seems, but easy to handle soft-wise.
Pendulums and gyros
Or we could get into real physics, and have fun with pendulums we’d put away from equilibrium with motors and that would have the sphere lean in a certain direction when they try to go back to equilibrium. Or use those unexpected torques gyroscopes create when they’re spinning and they’re pushed on in certain directions (those that tend to change their rotation vectors).
It’s likely we could do anything with that (i.e. rotate anyway we want), but it’s also likely there’ll be heavy math and heavy mechanics – do we want to get into that ? This deserves further investigation, as we’ve seen pretty cool things online. Stay tuned !
PS : Yes, Jim Rose should definitely be coach for the 2013 season.