This last season, I saw several drivetrains employing omni’s at one end of their tank drive including our robot. I see the advantage of this being you don’t need a center drop and can get better stability forward/backward without sacrificing forward traction since you have all wheels contacting the carpet. The downside is that it is appears easier to spin a robot defensively with omni’s at one end.
We had good maneuverability but I can’t honestly say it was better or worse than our traditional WCD with 1/8" drop center.
What were team’s experience with omni’s at one end? Do you plan to stick with them?
Thanks
Both are fine. I definitely would recommend drop center though for a few reasons:
-
It’s harder to get pushed around (when you have omnis in the back, it’s easier to get turned by other robots)
-
You get more traction with regular wheels (omni rollers have lower CoF if I remember correctly)
-
If you use single omni rollers, it would be a rougher rider than perfectly round wheels
It’s up to you at the end of the day. I’d go with drop center. Just my $.02
What about from an autonomous programming point of view; do omni-wheels make autonomous routines notably more consistent?
I would think that they would, since they would theoretically maintain a constant axis of rotation and reduce/eliminate slippage due to scrub forces.
From messaging 1678 (very responsive, by the way) this was one of the reasons they mentioned for doing omnis in the front.
We had tons of trouble getting our robot to handle consistently in 2017 with a dropped center six wheel drive, and almost no trouble in 2014 with a dropped center six wheel drive.
In 2014 we had aluminum wheels with tough blue nitrile tread, and quite a lot of center drop (3/16" I think?). There was almost no scrubbing when we turned, and tread wear was minimal.
In 2017 we used colson wheels with very little center drop (3/32" I think?). With fresh wheels there was very little scrubbing, but after a few matches the robot turned noticeably differently, as the center wheels wore down.
Just a couple data points.
There are two cases in which using either a center drop or omnis are clearly beneficial.
Center Drop is perfect for difficult terrain maneuverability over obstacles or up ramps. Center drops with pneumatic wheels were a vast majority for Stronghold in 2016.
Omni wheels are effective in scenarios where you want to manipulate the rotation point of the robot. The most common set up for this is the 2+2 (two traction, two omni) where the robot rotates between the traction wheels. This was common in 2017 with Steamworks for teams who wanted control when lining up for the peg, and a shorter travel time turning to face it. They are also a safe bet for a tippy robot, like in 2015 with Recycle Rush, but can be more finicky to drive.
For any other scenario, you are looking highly at minute details. I cover a lot of those details in this presentation I gave this summer if you are interested in diving deeper into it.
I’m a big fan of 6WD with omnis on the front from a design and maneuvering standpoint. I like not having a rock in the drivetrain. Also, less worry about certain wheels wearing down much more quickly than others. You also have more freedom in manipulating the center of turning than with a drop center. The concern about being turned more easily is valid, but in practice, there are so many situations that might favor either design, so it’s tough to say that’s a clear reason not to use omnis in the drivetrain. All in all, feel free to try it!
Looks like you guys need to rotate the tires every few thousand inches.
1 Like
It all comes down to your strategic assessment of the game each year. If you don’t need the specific mobility points already mentioned, go drop center with simple wheels. If you need that mobility enough to give up the possibility of being easily turned by other teams, use 2 (or 4) omnis. I would certainly consider 6 solid wheels with center drop the go-to default, but in the interest of total disclosure, 3946 has not competed with such a drivetrain except in our rookie year. Leaving out a now-indefensible decision our second year, we’ve done mecanum, H-drive, 10-wheel, and 4 solid wheel (short chassis). Each of those was driven by our strategic understanding that year, seeking (respectively) additional mobility, mobility, terrain capability, and agility (acceleration).
Honestly, it comes down to preference. I have seen some really bad drop center drivetrains. I have also seen some really bad omni drives. If it fits your strategy it doesn’t matter. Just make sure you make your drivetrain bulletproof. “But you can’t push with omnis yadda yadda yadda” , “you can get pushed with omnis yadda yadda yadda”, “you can push better with all tractions” honestly who cares about any of that. Just make sure you have a strategy and you execute it well, and build your drivetrain like a tank.
With that being said, we will most likely never run an all traction wheel setup again (2016 drives being the exception). Autonomous is a concern as is wheel wear and turning scrub (even though it is still good practice to replace your wheels every event). We will most likely run 2 Omni + 4 traction, or 2+2 in the future.