Before giving a talk at the university, I decided it would be a good idea to...spend 20 hours straight working at Sector on a omniwheel robot prototype. I only got a half hour sleep before the talk but I guess I must have been coherent based on the response from others.
See my slides on Google Drive at this link.
Here are some photos of the omni wheel bot.
Omniwheel before assembly of beads. The wooden block is a jig for mounting the press fit gears on the motors. The vertical mount design isn't great, as weight of the robot causes the mount to flex downwards a little, which pushes the gears together more tightly, which increases friction. That will be changed in the next version.
Some of the beads before mounting. I loaded them up on a piece of wire and coated them with Plastidip (black) or Flexidip (red, pictures on the full assembly below). It is better than nothing but still not a great solution. So far my best solution has been to skip tires entirely, and run the robots on rubber mats. I'm still looking for a good tire solution though.
The fully assembled omniwheel bot. I tested it with the power supply, and it does move. It draws a bit more current than planned (5 amps or so) but I think with a little redesign I can make a lower current draw version.
I got a bin full of more stuff donated to Sector, including some really nice motors to play with. The primary goal is still ultra-low-cost stuff, so I will start looking into sourcing cheap motors. My framework will come with practical notes, including 'buying guide' type notes for some parts, as well as design files, code, and everything else.