I have noticed a lot of teams in competitions using omni-wheels. I am the strategist for my team, and was wondering if it would make sense to slam into the side of robot to make them slide to the side when doing things like match loads. this reminded me that I have not slide proofed our robot. I was wondering whether it makes sense to add the four inch wheels with a high traction tire, instead of the back omni-wheels.
Based on our experience, I would advise not to use the high traction tires. Though they are called “high traction,” they don’t get much grip on the ground anyway. They definitely will help in preventing other robots from pushing you around, but the large tradeoff is slower drive speed (especially turning) and lower grip. We have tested the grip of both omnis and traction wheels to see which would work better in high stress situations. We found that the omnis were able to move the bot while the traction tires spun and kept slipping on the ground.
If your main goal is to not be pushed, this might be a good idea. However, be aware of this sacrifice you are making.
One thing our team is considering building our new chassis is using a combination, like omnis in front and traction in back. This would retain most of the advantage of fast turning from the omnis, but would still make it harder for bots to push us sideways. The primary issue that we’ve been considering with this system is that it would make your center of rotation very near the traction wheels, so unless you already have a large imbalance, it would really screw up turning.
Dropcenter drives?
Yeah, tell me how that turns out 
Alright, now get it over the 10 bar
We had something similar at the start of the year, it got over no problem.
Really? I’m genuinely curious now, how much vertical space was between the center and outside wheels? I would think the middle wheels would get elevated and lose contact with the ground
setup: 4" omni outside, 4"traction inside. Robot weighed roughly 20 pounds. Scissor lift, all steel (we were broke at that point). 6 motor speed drive, chained together. Everything was double redundant, ie any length of chain would snap and all wheels would have power. Any remaining length of chain snaps and all wheels still have power (I know, I’m extra.) Going over the bar, the middle wheel would lose contact with the ground as the front wheels went over, then the bar would get caught between the middle wheel and the front wheel, but because friction we managed to get over the bar, and the last wheel was either slightly on the bar or remaining in the 5pt zone. Definitely a lot of jerking around the mogo, but if the stack was stabilized and your mogo intake could place over the 20pt, it was pretty good. I think something else that was mentioned to me was putting the middle wheels as smaller wheels (2" traction or something), but have them mounted on a bar that could swing forwards/backwards (drag ur pencil over a paper, then put your finger in the way. The pencil will move back and up), so that we could clear the bar, but in normal match play would prevent us from being pushed sideways.
We ended up not doing that not because I didn’t want to but because we rebuilt, and uh…the rest of the team did … minimal work…
I have this configuration of wheels on my drive train, and I can easily put a mobile goal in the 20 point zone
Would it really screw up turning? I would make turning different than with four omni-wheels for sure. But I doubt you would hear from forklift drivers, for instance, that turning with a forklift is all screwed up. It’s not exactly the same wheel set-up, but the turning behavior is similar.
Also consider cars. If they turned on their centers, there would be things you could not do that you can as they currently turn. And currently they turn differently in forward v. in reverse, and you can take advantage of that if you know what you’re doing.
While I’m glad to have my video cited, the other concern with a drop center drive is that it by nature must tilt forward and backward a little bit. I would be concerned about that when making tall stacks or trying to line up with the stationary goal, especially as the center of gravity of the robot shifts during its motion. No problem in Starstruck (we kept it all season), big problem in ITZ. Also, as others have mentioned, I have no clue how a drop cetner drive would be over the starting bar. You could very easily get stuck on your four unpowered wheels, I believe. It seems like a no-go for this season at least.
Power the 4 outside wheels, and put the middle wheels on something springloaded, so that going over the bar they can give-way, but otherwise in normal match play serve as your center of rotation. I’ll build an example sometime soon (after march break).
Springloaded with enough force to hold up half the robot? That isn’t going to be able to be pushed up while going over the bar.
It’s easy to stop something from getting pushed up too far just by putting something non-flexible in its way when it hits a limit. Note the bigger key to what @ranOOm said: all four omni-wheels are being powered. So to lose power to drive at all, somehow all those wheels must come off the floor, not just the middle ones.
But I worry about the rocking you mentioned above. When you’re trying to do things like lines up with horizontal bases to mobile goals or to vertical stacks, it’s nice for your robot to remain aligned with the vertical and the horizontal.
I think I was confused because what @ranOOm is suggesting sounds like the opposite of a drop center drive. In a drop center drive, the middle wheels are the only powered wheels, not the only unpowered wheels.
Ya, it’s not what I was thinking specifically, either. I don’t think @ranOOm is saying they’re not powered, though, but maybe he is. Either way, it does accomplish having the traction wheels on the floor to avoid slipping without totally messing with the likely center of rotation without them. I would think getting the right strength in the suspension would be tough, but doable.
I would be careful about high traction on the back wheels. I used 3.25 in omnis on the front wheels and then 3.25 in high tractions on the back wheels. While unloaded the robot turns fine however once you have a sufficiently large stack turning becomes a nightmare as the center of gravity is the front of the robot where the mogo is located but the traction wheels make the center of rotation at the back edge of the robot. Turn speed is greatly reduced even with a 6 motor drive.
Why didn’t you switch the wheels? Put both center of gravity and rotation at the front.