Mecanum Drivetrains

[yash101]

Originally Posted by Pendulum^-1

Limitation number one: Mecanum wheels are very easy to push. A problem on a field with well-defined chokepoints, such as the 2013 arena.

I have to disagree with this. We were able to easily push around the other robots. We used four CIMs, one on each wheel.

Even if you are being pushed around, the maneuverability of mecanum allows you to typically free yourself easily.

Mecanums are quite a friendly drivetrain if you want to have full control on how you move, without getting way too complicated. It will go on just as regular wheels would, however, using a separate transmission for each wheel. The rest of the changes required are all in the robot code. Also, if your mecanum robot code isn't ready, your standard code will work, only without support for the advanced features like sideways driving

[cmrnpizzo14]

There are enough teams in the FLR area that you can find some easy resources right in town. 3173 has used mecanums this year and 2011. 191 has used it prettymuch every year I believe. 1551 has used mecanum or octocanum.

Honestly, we enjoyed our experience with it but we felt that we could accomplish equal or better driving with a simple kitbot on steroids. We prototyped it over the summer and loved it.

Tips:

-Remember to have your wheels in the "X" configuration as opposed to the "O"
-Nanotubes make construction very easy and helps save the hassle of chains but they are heavy and rather difficult to get to the bolts. If these are used, purchase spares for competition so that you can simply swap the whole nanotube instead of trying to pull one wheel off. We have found this to be faster.
-Keep the frame square but flexible. We found that this was difficult this year. If a frisbee became jammed in our hopper the best method to free it in match was to ram the wall. This freed the frisbee (most times) but unfortunately would mess up our nice square frame because we could not keep it perfectly rigid. We had a strategy this year that was not too reliant on driving so it wasn't a huge deal but our chassis team did have to resquare the frame after many matches.


For fun, ask 578 about their mecanum tank treads that Eric D. created. Inefficient, heavy, largely useless, but very cool.

[EricH]

Originally Posted by yash101

It will go on just as regular wheels would, however, using a separate transmission for each wheel. The rest of the changes required are all in the robot code.

This is a HUGE benefit of mecanum if it's your first time doing it. You can even (should you build a practice robot) build for an easy switch. 330 did that in 2005--never did put the extra trannies onto the competition robot, though; it stayed as a 2-gearbox, 6WD despite having the ability to go mecanum.

One thing I haven't seen brought up, but I've seen on a mecanum drivetrain (not on the competition robot for that year, just a test mule): If you have a heavy-ish load located well outside your frame perimeter, don't expect to be able to strafe straight without closed-loop control. This goes back to the whole "keep all 4 wheels on the floor" principle, but more along the lines of "keep the same weight on all 4 wheels" to the best extent that you can.


Oh, and my opinion of mecanum is this: If properly executed, and in a game that actually requires the extra mobility, and properly used, it's a pretty big advantage, until you run up against a swerve. The swerve is better in terms of traction, and worse in terms of complexity. If, however, the mecanum is not properly executed, or used properly, or the game doesn't actually require the extra mobility, then the team could have probably used the weight better by building a 6WD (or similar tank drive) and allowing for more material elsewhere in the robot. Probably. (Note that this does not apply if "cool" is a driving factor in the design more than "competition effectiveness" is, or to any offseason robot built for non-competition purposes.)

[pfreivald]

Originally Posted by cmrnpizzo14

1551 has used mecanum or octocanum.

We've used octocanum the past two years, and mecanum three others--and we're always upping our capabilities, especially with control. One thing I have my programmers working on this fall is using the gyro to compensate for wheels not working the way you want them to because you're driving over an irregular surface.

It's...interesting, anyway!

Either way, if you want to come down and check out what we've done in more detail, let me know and we'll set something up.

[Chadfrom308]

Our team in 2012 used mecanum wheels with regular traction wheels. The mecanums were on an offset axle that got pushed down with pnumatics. We were able to drive really well and place some nice defense.

I will say though, you have to be careful. Mecanums are can be dangerous. We got stuck on a ball one match and literally tore it to shreds in seconds. We also ran over team 67's go pro and spun our mecanums on it by accident. Luckily it was okay! I wish I could've saw that video though...

[pfreivald]

Originally Posted by Chadfrom308

I will say though, you have to be careful. Mecanums are can be dangerous. We got stuck on a ball one match and literally tore it to shreds in seconds. We also ran over team 67's go pro and spun our mecanums on it by accident.

Is that more dangerous than any other drive train?

Also, I would like to see that video, too!

[Brandon Zalinsky]

Originally Posted by Akash Rastogi

This sounds pretty creative! Do you have pictures or videos of it in action?

Here's the design and implementation of the Locking Mecanum Drive 1058 built in 2011. I threw in some match footage for you guys too!

With regards to what Jay said about Field-Oriented Drive, here's a video for that.

[Madison]

Originally Posted by Flak-Bait

Here's the design and implementation of the Locking Mecanum Drive 1058 built in 2011. I threw in some match footage for you guys too!

With regards to what Jay said about Field-Oriented Drive, here's a video for that.

If I'm understanding this correctly, the rollers are stopped only be the pin / screw / post that goes through the 2nd hole you drilled near each rollers' axle?

[BrendanB]

1058 is definitely one of those teams that has re-defined how to use mecanums effectively in the game. Their 2010 robot with the field oriented drive performed amazingly throughout the 2010 season. It was fast, responsive, very agile, and played some very mean mid-field offense!

I didn't get to see too much of their 2011 robot but I've heard a lot of stories about what they pushed.

Their 2007 and 2008 mecanum drives were also effective players of the game.

While mecanums still aren't my personal choice to use in a drivebase they can make for a great off-season project for a team. Plus they make excellent demo robots.

Use in a competition is a different story but every team needs to make decisions when they design their robots. There have always been times where being able slide sideways is advantageous and mecanums are a very easy way to achieve it. There are downsides that go with them but there are downsides of every mechanism you put on your robot.

In the end, no matter what drivebase you go with the best thing you can do is train your drivers and give them time to practice under realistic circumstances. While our team doesn't use multi directional drives we've trained our drivers to react faster so they can quickly maneuver around obstacles without us giving them another direction. We use old robots, chairs, trash cans, simulations, etc to keep them on their toes so when their path is blocked they don't stop moving while maneuvering.

Practice, practice, practice and the decisions you make at the beginning of the season will pay off no matter what drivebase you use.

[Brandon Zalinsky]

Originally Posted by Madison

If I'm understanding this correctly, the rollers are stopped only be the pin / screw / post that goes through the 2nd hole you drilled near each rollers' axle?

In the first shot in the video, the new hole that is being drilled is to help hold on a piece of rubber. The wheel itself is left unmodified, and we take a third mecanum plate and put one piece of rubber on each "wing" or flange of the plate. When the piston is activated, it pushes that aluminum bar you see in the video into that third plate, which is mounted parallel to the mecanum wheel. This pushes the rubber piece into the wheel, braking the roller.

The plates are on the same axle as the wheel and are spring loaded, so when the piston retracts, the plate comes off the wheel, unbraking the roller.

[Akash Rastogi]

Originally Posted by Flak-Bait

In the first shot in the video, the new hole that is being drilled is to help hold on a piece of rubber. The wheel itself is left unmodified, and we take a third mecanum plate and put one piece of rubber on each "wing" or flange of the plate. When the piston is activated, it pushes that aluminum bar you see in the video into that third plate, which is mounted parallel to the mecanum wheel. This pushes the rubber piece into the wheel, braking the roller.

The plates are on the same axle as the wheel and are spring loaded, so when the piston retracts, the plate comes off the wheel, unbraking the roller.

I have to admit, that is a great little feature! How well did it work out for you in a real match against defense?

[Brandon Zalinsky]

Originally Posted by Akash Rastogi

I have to admit, that is a great little feature! How well did it work out for you in a real match against defense?

Very well, even though we encountered very little defense focused solely on us. We'd do a "roll", where we'd activate the traction drive, plow into the side of the defensive robot, go back to mecanum mode, and push towards the goal while rotating the robot. This would cause us to roll off the side of the defending robot and get past them. In other situations, we could just outdrive them with the agility of the mecanum wheels. The biggest advantage of the locking mecanums is if an opposing robot is pushing you sideways, you can stop dead with a push of the button and, amusingly, confuse the hell out of the opposing robot's drivers.

[yash101]

Team 1165's Mecanum Test! Mecanum is nothing but pure amazing, and it actually seems like magic, especially with the wheels covered up so no one can see them
http://www.youtube.com/watch?v=_Lrosi-WA7A

[pfreivald]

Originally Posted by Flak-Bait

Very well, even though we encountered very little defense focused solely on us. We'd do a "roll", where we'd activate the traction drive, plow into the side of the defensive robot, go back to mecanum mode, and push towards the goal while rotating the robot. This would cause us to roll off the side of the defending robot and get past them. In other situations, we could just outdrive them with the agility of the mecanum wheels. The biggest advantage of the locking mecanums is if an opposing robot is pushing you sideways, you can stop dead with a push of the button and, amusingly, confuse the hell out of the opposing robot's drivers.

A lot of the same advantages of octocanum, though with octocanum you can do a 5:1 or greater downshift with little weight and cost built in.

[Lil' Lavery]

Being able to execute a "roll" maneuver is pretty much the tempts me towards omni-directional drivebases (and some fine tuning near goals like 2005, 2007, or 2011). That being said, I don't anticipate it happening anytime soon on 1712 and it's only useful if the field has enough space for it to be executed (so it rarely solves choke point defense). 1640's swerve drive roll maneuver is probably my favorite, though. So beautifully executed this season.