We are using NEO’s and 4" omni’s because we were 15lbs over weight. With mini-cims and 6" omni, drive pods were about 5.5lbs a piece. Our chassis was about 20lbs. So total drive train weight is about 65lbs. Which is a little heavy. With Neo’s and 4" omni’s, it’s about 50lbs.
It’s better than our 8 mini-cim mecanum by about 20%. Plus with that many wheels on the ground if one wheel loses traction, we don’t loose complete traction like mecanum.
Do you have any video? We’ve been working on a differential swerve design during the off season and didn’t think we could get it ready in season.
This was from a few weeks ago, we used this for premier night.
That is awesome. Did you have problems tuning the PID? We found that to get speed, we had a very low gear ratio but that resulted in a super squirrelly behavior when swiveling. We couldn’t get it to now overshoot or undershoot. Are you using any velocity profiling/following on the swerve to get it to behave better? It looks super well tuned.
As far as I can tell you are the first to run differential swerve in competition. Congrats.
Good luck on your season.
Thanks for the kind words. The team has spent a LOT of time tuning. They tried a lot of different approaches and but it finally came together. Very proud of these students and what they can accomplish. We will be releasing the tweaks to the drive modules as well as the code for what we did as soon as we have time to document and publish. We started with our custom swerve code we developed over the last 3 years.
We are still working on the acceleration and deceleration curves. With 8 Neos and that much torque we have to slow it down a lot to keep from pulling those wheels off the ground.
Good luck at St. Louis and Smokey Mountains. I’ll be eagerly watching.
We used 8" AndyMark mecanum with 10.7:1 Toughbox mini gear box & NEO motors. Competitors during playoffs focused defense on us, but we were still able to score hatches & cargo, while our alliance captain team 346 loaded cargo & placed hatches at will. The mecanums gave us great mobility around the field & were great aligning up on targets. Team 1262 the Stags
Ran our first regional last week. Here is an example of some pushing going on. We can’t plow thru just anyone but we can hold our own against a typical KOP bot.
Here is the video:
This year 1108 went out on a limb and designed our own drivetrain.
We called it “Panther Drive.” Was good for a Creativity award, but did not perform well and our season is over. But we had a great time and learned alot.
We put a mini-cim inside the wheel, used two Andymark 20-dp ‘planets’ and 3d-printed our own ring gear using CF-20 carbon fiber.
We put this motor on a steerable yoke, put one in each corner and made a fully-strafeing double-crab-drive robot. Our two front wheels steered about 270 degrees, our two backs did too.
It was a bold plan.
I first saw a motor in the wheel in 2011 with PWNAGE, but they did a two-pass gear system. and a much larger wheel. They haven’t done it for years now. I’ve never seen another. Except all the bicycles in China today have hub-motors.
Ours was smaller than PWNAGE, so we could get great acceleration with 4" dia wheels set up for about 20 fps. Originally planned to do it as a wheel in a regular six-wheel drop center tank drive. Other mentors on 1108 encouraged going to steered modules.
No chains, no belts, no transmissions. Clear space in the chassis. Center of gravity is really low. Potential for spring suspension too. Efficiency. All of those things rob power and take up space.
Originally designed to be 3d printed, they worked okay that way, but when the motors got hot, the PLA sagged. Had the parts made in aluminum after bag-day and switched out. In 3d printed, we had a module on each corner of the robot for $94 each. Still a really good value with machined parts, estimated $200/wheel, not counting steering motor.
We didn’t have a failure of a gear-set or wheel, but two times welds broke on our aluminum steering yokes to leave us without a wheel and with a shorted motor controller. We have a plan to do those welds better.
The vast majority of our failures were related to the software and the crab steering. We sat still for far too many matches in two regionals for anyone to consider picking us.
Team 1899, Saints Robotics, did swerve drive, but they have low gearing to enable them to play amazing defense. Their swerve is probably the strongest out of almost every PNW team.
4 omnis on the corners and 2 wide colson wheels and two thin colson wheels and a octagon frame
Did you design it? Pic?
I’m on team 2522, but they had played defense against us at PNW Glacier Peak: QF 3 and 7.
how heavy is the drivetrain with bumpers
I think we are somewhere between 106 and 110 with bumpers but with out bumpers we were 96.6
Strongest by pushing power? In my opinion, they’re very evenly matched with Jack In the Bot.
1899 used AndyMark swerve. There’s a fantastic video of their swerve (CIM and 775pro) up against Jack In the Bot (double NEOs) that shows the difference in speed and pushing power.
if you change those wheels to mecanum, you’re golden.