pic: Vex Swerve v3

[JVN]

loading photo ...

[Lil' Lavery]

Way to steal my thunder John, way to steal my thunder
Way cleaner and more compact than the modules I'm prototyping (mainly due to the custom-ness of the brackets).
A few questions though. What are the metal plates above the corner gears? How are the swerve modules rotated? Why omni-wheels?

[Nuttyman54]

Originally Posted by Lil' Lavery

Why omni-wheels?

I wanted to know that as soon as i saw the pic. I'm gonna make a wild guess and say that it's possible to rotate the wheels to make it holonomic? that's the only reason i can think of.

[Pat McCarthy]

Wow, very nice. You should work for VEX or something.
Oh, wait...

[Donut]

Behold the power of Vex.

I'll join the chant; why omni-wheels?

[crazykid234]

I too tried to do something like this, but I had a problem with the modules flexing to the sides. How did you avoid this?

[JVN]

It is 2 omni-wheels, 2 traction wheels.
This means when it isn't swerving, it skid-steers smoothly.

The metal plates on top of the gears are used with a washer as part of a custom thrust bearing that the swerve module rotates on. This results in a high amount of module stability, while minimizing rotational friction.

For more pictures, and some videos:
http://www.vexlabs.com/video/swerve/

[Vexguy365]

Way to go. I had a crab drive VEX 'bot that worked OK. All of the wheels weren't held in position tightly enough so they had a little wiggle.

[Alex Cormier]

Originally Posted by JVN

It is 2 omni-wheels, 2 traction wheels.
This means when it isn't swerving, it skid-steers smoothly.

The metal plates on top of the gears are used with a washer as part of a custom thrust bearing that the swerve module rotates on. This results in a high amount of module stability, while minimizing rotational friction.

For more pictures, and some videos:
http://www.vexlabs.com/video/swerve/

once again i am simply amazed at what you have made and what vex is capable of.

[Ben Piecuch]

John,

I see that you used quite a few gears to increase the turning speed of the crab modules. I'm assuming that directly driving the modules was too slow? And, when you increased the speed (thereby decreasing the torque) you found that you required a 2nd motor to handle the job?

Very elegantly done. Between the two-speed, 9fps bot another student build, and now your crab-drive, I'm going to have to find something else to motivate little Art in this off-season!

Can we expect these "custom crab module plates" to be available anytime soon? Great job as usual, John!

BEN

[JVN]

Originally Posted by Ben Piecuch

John,
I see that you used quite a few gears to increase the turning speed of the crab modules. I'm assuming that directly driving the modules was too slow? And, when you increased the speed (thereby decreasing the torque) you found that you required a 2nd motor to handle the job?

Ben,
The module steering is done with Vex servos. (This allows for a "no-programming" swerve drive.) These servos have LESS rotation than I wanted my wheels to have, so I geared them up (not to increase the speed, but to increase the range of motion). I used the large 84 tooth gears simply because it created an elegant package with a minimum of gearing. (The less reductions used to transmit the steering torque, the less slop your modules will have; using chain isn't an option, because the chain stretched too much.) Once I used the basic 84:60, I realized that was too much increase, so I did some preliminary slowing. (I broke the rule of never slowing something down, only to speed it up again).

The resulting steering ratio was:
60:36, then 84:60.

Originally Posted by Ben Piecuch

Can we expect these "custom crab module plates" to be available anytime soon? Great job as usual, John!

BEN

I think there may be a "swerve" kit available shortly.
Believe it or not, I don't just sit around playing with robots, this is a new product. I only hope there is some interest.

JV

[GeorgeTheEng]

Originally Posted by JVN

I think there may be a "swerve" kit available shortly.
Believe it or not, I don't just sit around playing with robots, this is a new product. I only hope there is some interest.

If you did get to "just sit around playing with robots", IFI would be receiving hundreds of resumes a day from FIRST alumi, mentors, and engineers (mine included probably).

That is a pretty impressive mechanism. It looks like it works well. My only comment would be that this begs for some sort of 90 deg gear transfer. When I watch the video I get a little concerned about the motors on the swerve drive when they swing outside the dimensions of the body. If they could be mounted above and transfer rotation through a bevel gear or something.

George

[Nuttyman54]

Originally Posted by JVN

I think there may be a "swerve" kit available shortly.
Believe it or not, I don't just sit around playing with robots, this is a new product. I only hope there is some interest.

JV

I'm SOOOO getting one of those when it comes out!

[Lil' Lavery]

Originally Posted by JVN

Ben,
I think there may be a "swerve" kit available shortly.
Believe it or not, I don't just sit around playing with robots, this is a new product. I only hope there is some interest.

JV

Booo, we were gonna exploit the fact that there wasn't a swerve kit and teach teams how to build a swerve drive once we perfected ours, lol.

I'll post pics/vids of ours (built with standard parts) once we do some more testing. (The first version that we have now is purely to test the swerve module design itself, next we will have a semi-swerve drive built out of a single vex kit, then a full fledged crab drive).

[henryBsick]

Is there any reason you placed the VEX robot controller on the the same side as the servos?