Improved Mecanum Wheels

[Adam Richards]

Originally Posted by dlavery

Oh? Why?


-dave

Dave, don't be mean to people who haven't already read the technical specs for this year's water game. It's not our fault we aren't on the game design committee.

[dlavery]

Originally Posted by Adam Richards

Dave, don't be mean to people who haven't already read the technical specs for this year's water game. It's not our fault we aren't on the game design committee.

Admittedly, a bit tongue-in-cheek. But not entirely. There are multiple layers of real questions embedded in that simple little two-word interrogatory. Not all questions, and not all data, are obvious.

-dave

p.s. if you reacted quickly like some and think that the original question means "hey, don't use Mechanum wheels this year" then you haven't looked deep enough.

[Cory]

Originally Posted by dlavery

Admittedly, a bit tongue-in-cheek. But not entirely. There are multiple layers of real questions embedded in that simple little two-word interrogatory. Not all questions, and not all data, are obvious.

-dave

I think what Dave is trying to say is that it might not be the best decision to arbitrarily choose what drive system you want to use, before having any clue what the robot needs to do.

[mechanicalbrain]

Originally Posted by Cory

I think what Dave is trying to say is that it might not be the best decision to arbitrarily choose what drive system you want to use, before having any clue what the robot needs to do.

I agree, but I think we'll stick with our mecanum propellers for this years water game! Seriously though, what are the major disadvantages? My team is looking at these and I figure ill take advantage of other's experiences.

[Cory]

Originally Posted by mechanicalbrain

I agree, but I think we'll stick with our mecanum propellers for this years water game! Seriously though, what are the major disadvantages? My team is looking at these and I figure ill take advantage of other's experiences.

Not much of a point in using them if the game doesn't require manueverability.

Or if there are obstacles that you need to climb.

They won't necessarily be bad for a given game.

But why build a swerve drive if the kitbot will serve your purposes just as well?

[Alan Anderson]

Originally Posted by mechanicalbrain

Seriously though, what are the major disadvantages? My team is looking at these and I figure ill take advantage of other's experiences.

I can think of two major disadvantages, one mid-level technical issue which can range from trivial to showstopper, and quite a lot of little details.

The disadvantages: some of the mechanical power available from the motors is lost to sideways motion of the rollers, so the robot won't be able to push as hard; and it doesn't work well at all unless all four wheels are solidly in contact with the ground.

The technical issue is the control system and software. That's very easy in concept, but might quickly get more difficult depending on how good your programming and system engineering can be.

Details of wheel size, roller mounting, speed sensors, traction, drivetrain, etc. will probably make doing it more of an art than a science unless you've done it before.

[Veselin Kolev]

I for one am willing to try these wheels out. Though there is the obvious disadvantage of having a less efficient system of transferring power to the ground, you are given an almost infinite degree of maneuverability. You can drive forward or backward, arc forward or backward, drive sideways, arc sideways, and strafe at any angle, just like swerve drive. It is like swerve drive, but a lot simpler. I think it is a very good drive train for FIRST teams to try out. The programming is tricky but not impossible for rookies, its really just figuring out the math behind the motion. The fabrication is easy if you figure out the tricks behind making it simple.

The only thing you have to consider is that it limits your driving strategy. With this drive, you are not going to be able to win pushing fights against some robots. Robots with six-wheel drive and a low speed are going to ram you into a corner if you have this drive. The point of this drive is for you to make it, and program it, and get the driver so used to it that it is a breeze to fly around such slow moving robots, not ever touching them. This drive is one that demands a coordinated driver, and a fast gearing, maybe 12 feet per second or more. That way you are a super-agile monster of a robot that does tasks fast and never touches another robot.

But of course, I’m just a proponent for this drive train. It all really depends on your driving style. This is just something I’m interested in seeing on the field.

[EricH]

Only one problem with whizzing around those "slow" six wheelers. If they get between you and your objective(s), one "slow" robot, driven right, can block a mecanum. We actually tested this out, and it was one reason we chose not to use mecanum this year. Our mecanum Kitbot could not get around or through our '03 bot by any means, and '03 is not easy to steer (though we didn't need to to block Kitbot). Also, you need your CG pretty well centered, or your robot will not get the full mecanum effect. Moral of the story: TEST FIRST! It could save you a lot of pushing and other fun stuff at competition.

[Cory]

Originally Posted by Veselin Kolev

It is like swerve drive, but a lot simpler.

So is a holonomic platform.

And it's a lot easier mechanically, particularly now that we have the NBD conversion, and AndyMark for the omnis.

I imagine the programming is along the same lines of difficulty as with mechanum.