Slide Drive

[Frenchie461]

I was looking up stuff for this year's Vex game, and I came across a drivetrain similar to nonadrive, where there are 5 omni wheels in a H shape, and I was wondering if it had ever been used in FIRST? It seems like it would be a simple method of omnidirectional movement that had some pushing power.

[Hawiian Cadder]

we attempted it this year. in order for it to have more pushing power than a normal drive you really need either 5 cims, or to power the middle wheel with some of the other motors. the one thing I really liked about it was the ease at which it could be changed to 6 wheel tank I beleive 148 did something like that in 2010, a sort of slide drive + tank drive + nonadrive.

[AdamHeard]

51 made it to einstein this year with it.

[xSAWxBLADEx]

yes it has, we did it in 2004 maybe 2003 i dont remember the exact year. It was before me

[Vikesrock]

2175 used it this year. Based on their experience I would recommend mounting the center wheel using some sort of suspension unless the floor is perfectly flat. The bumps around the towers would lift the strafe wheel off the ground when it was rigidly mounted.

They used 4 CIMs on the outside wheels and 2 775s to power the strafe wheel.

After coaching 2175 at Champs I have mixed feelings about the drivetrain. Like other omnidirectional drivetrains I would recommend a practice robot or other way to get substantial drive practice in order to take full advantage of the increased maneuverability.

I personally would have preferred a standard 6wd or 8wd setup.

[Colin P]

I've only ever heard it referred to as "H Drive", and we were contemplating using it this year but just went with mecanums.

[BrendanB]

Originally Posted by Vikesrock

After coaching 2175 at Champs I have mixed feelings about the drivetrain. Like other omnidirectional drivetrains I would recommend a practice robot or other way to get substantial drive practice in order to take full advantage of the increased maneuverability.

I personally would have preferred a standard 6wd or 8wd setup.

Great point! A lot of H drive and mecanum teams don't get the practice they need and mostly drive their robot as a tank rather than using their drivebase for what it is made for.

[Chris Hibner]

Originally Posted by Vikesrock

2175 used it this year. Based on their experience I would recommend mounting the center wheel using some sort of suspension unless the floor is perfectly flat. The bumps around the towers would lift the strafe wheel off the ground when it was rigidly mounted.

Even with the (relatively) flat floor this year, having suspension on the center wheel was a necessity. We were always amazed at how much the center wheel floated up and down even when we thought it was on a perfectly flat floor.

[Aren Siekmeier]

Originally Posted by Vikesrock

2175 used it this year. Based on their experience I would recommend mounting the center wheel using some sort of suspension unless the floor is perfectly flat. The bumps around the towers would lift the strafe wheel off the ground when it was rigidly mounted.

They used 4 CIMs on the outside wheels and 2 775s to power the strafe wheel.

After coaching 2175 at Champs I have mixed feelings about the drivetrain. Like other omnidirectional drivetrains I would recommend a practice robot or other way to get substantial drive practice in order to take full advantage of the increased maneuverability.

I personally would have preferred a standard 6wd or 8wd setup.

At the beginning of the season, I also strongly advocated a 6wd drop center. Most people argued that strafing would greatly increase tube placing speed, but I was of the opinion that practice would make tank far superior. We went with the slide drive, in the configuration Kevin described, and I was pretty happy with how it drove after we added that awesome racquet ball to shock mount the strafe wheel. Even on the flat areas of the field, you couldn't guarantee 5 wheel contact (only 3 of which define a plane).

And of course, even with the added strafing that supposedly made tube scoring faster, it was still just as slow because we simply did not practice nearly enough. We could have done some pretty sick stuff with it (which, granted, would have hardly been necessary), but the lack of practice just made it unwieldy in the driver's hands, much like an unfamiliar swerve drive.

I think its definitely better than the typical four wheels in a circle set up, since the forward direction is almost always favored and the strafing is an after thought. It adds a few motors, but it's quite simple to add the fifth to a typical tank drive setup. The amount of practice involved seems to me to be a wash, since for both you just need a lot. The omni wheels do have a surprising amount of traction and you can at least hold your ground, even if you can't push people out of your way. If traction weren't an issue at all, I would probably go with it every time for the maneuverability. But that's hardly the case (unless it's fast enough and your driver is good enough to get around the defence every time).

Our robot's CAD is posted at http://www.chiefdelphi.com/media/papers/2556 and shows the slide drive configuration. Unfortunately, the awesome racquet ball that we added later in the season was one of those details that slipped through the cracks and didn't make it into the CAD.

[Jeremy Germita]

My FTC team (72) used a slide drive base(inspired by FRC148 in 2010, sans traction pods) up until our first competition. (The video linked above in Techhexium's post).

Personal experiences:
As driver, I found that it helped a lot with lining up for scoring, but we did have some traction issues. The mountain in the middle of the FTC field was difficult to traverse with these wheels. Seeing as though VEX Gateway's field is flat, this may not be a problem for you.
In the end, we scrapped the drivetrain after deeming it unsuitable for our strategy of traversing the obstacles.

As with the choice of drive train(or any system, for that matter) in any robot competition, there are a series of pros and cons.

Pros:
Maneuverability - 3 degrees of freedom: X Translation, Y Translation, and Yaw vs 2 degrees(Y translation, Yaw) of a typical skid steer drive
Faster scoring in autonomous mode - The robot is not spending too much time turning and driving... turning and driving... etc, just translating along the X/Y plane and only turning when needed.


Cons:
Loss of traction - omni wheels of any kind typically have a lower Coefficient of Friction wheels
Complexity - More motors, transmissions, etc
Weight(only a real issue in FRC) - See complexity


Regardless of platform, you also have to worry about your chassis flexing. You might remember from Geometry class that between any three points, there lies exactly one plane. With Vex or Tetrix metal, it is relatively difficult to make all 5 wheels(left 2, right 2, strafe) touch the ground. Depending on where the flex is, you might have trouble strafing or translating across the field. Also, on that note, weight distribution is key to ensure the robot moves as desired. For example, if more weight is on the two front wheels and less on the back and strafe wheels, the other wheels may begin to slip.
I highly recommend implementing a form of speed control on the wheels. Search CD for threads about speed control and mecanum drive trains if you are interested.

Also, like all drive trains, slide drive could only be used to its full potential with practice. No use giving it a strafe functions if you drive it like a tank, right?

If you have any more questions about FTC072's experiences with the slide drive system, you can PM me.

[Techhexium]

In FTC, a team used a slide drive for their robot. Not really related to FRC (FTC robots use either standard wheels or ommi-wheels, no mecanum) but can be related to VRC. The team have videos of two iterations of their slide drive here