pic: FTC 4977 - The Black Box

[Andrew Lobos]

[cdm-description=photo]38256[/cdm-description]

[CalTran]

And here I was thinking that an FRC type robot would be impossible in FTC. Scale that thing up a bit, rotate the roller claw 90°, and you probably could have done well in either Rack N Roll or Logomotion. Impressive.

[Peyton Yeung]

Can we get a video soon?

It looks so awesome and I'd love to see an FTC Swerve:)

[Andrew Remmers]

I'm glad people are finally doing swerve drives and crab drives. Seriously I've been trying to get mind competition ready for about 3 years now. Never got the chance to do it fully.

- Andrew

[CalTran]

Mind if we get another angle of just the swerve? If I'm seeing it correctly, looks like you pulled off a coaxial swerve in Tetrix parts.

[Andrew Lobos]

Mind if we get another angle of just the swerve? If I'm seeing it correctly, looks like you pulled off a coaxial swerve in Tetrix parts.
Yep, it's all Tetrix parts. The swerve modules were our summer project, and we designed them to the 2011-2012 rules. It was originally independently driven and steered, with servos as the steering motor, but the rotation response was so awful we switched to motors.

Here is a photo from another angle: http://i.imgur.com/66Ube.jpg

[Wayne TenBrink]

Very nice! How do you get the rings off the dispenser? How do you release the rings on the peg?

[CalTran]

Very nice! Is that a roller claw? How do you get the rings off the dispenser?

Probably position the claw under them and suck it in

[Andrew Lobos]

Here you go: http://i.imgur.com/3y058.jpg

The swerve modules are all Tetrix components. The modules were our summer project, and were build to be legal with the Bowled Over rules. We're working on a revision that takes advantage of the new part allowances in the new rules

Also, I made a post with this image earlier and it didn't seem to show up... sorry if this becomes a double post.

[Andrew Lobos]

Very nice! How do you get the rings off the dispenser? How do you release the rings on the peg?

In autonomous, we have a small flap that folds out when the robot drives forward. All of the rings get pushed onto the ground and we pick up off the ground.

For releasing the rings on the pegs, we just pull the lift down while the ring is on the peg.

Some more detail on the roller:
The belts themselves are McMaster "flat urethane belting". The surface is fairly slick, so we ended up coating the rollers with tape. This was mainly a "quick fix" to get us through the first competition. We plan to find a more long-term option that holds the rings better.

[maxweberh]

Was I the only one thinking they meant an actual black box? -In aviation, a "black box" (they are actually bright orange, to facilitate their being found after a crash) is an audio or data recording device in an airplane or helicopter. The cockpit voice recorder records the conversation of the pilots and the flight data recorder logs information about controls and sensors, so that in the event of an accident investigators can use the recordings to assist in the investigation

[CalTran]

In autonomous, we have a small flap that folds out when the robot drives forward. All of the rings get pushed onto the ground and we pick up off the ground.

Curious. How does your alliance partner take to that? Do you clear all the rings or only certain rings?

[Andrew Lobos]

Curious. How does your alliance partner take to that? Do you clear all the rings or only certain rings?

For most partners, it's fine since we only clear one "peg" on a dispenser in autonomous. If the partner insisted on having both pegs of the dispenser full on our side, we went to our other dispenser and cleared that one in teleop.

In the future, we may consider changing to a "bucket" style pickup but for our next competition we're sticking with the roller claw.

[CalTran]

Fwiw, I think that going with a roller claw, and extension to active manipulation, is brilliant and will set you apart from other teams. Active manipulation (read: floor pickup) in this game, from what ive seen, is rare but allows you to not be limited to just what's on the pegs and allows multiple tries per ring, if needed

[Andrew Remmers]

That module is HUGE, but I like it. Did you guys do any CAD at all? Also any video?

- Andrew

[Andrew Lobos]

Here's a video from one of our matches.
http://www.youtube.com/watch?v=8Gk5cjU45MA

The PID tuning on the rotation motors is still a bit rough, which causes some of the left and right misalignment you see.

[5383robotsrule]

what language did you guys use to program this ? and is there any cad of the swerve modules ?

[Andrew Lobos]

what language did you guys use to program this ? and is there any cad of the swerve modules ?

We used RobotC - the code for this robot is available on GitHub at https://github.com/FTCTeam4977/Robot2012-2013

I'll talk to the mechanical guys about getting the CAD up - the modules were CADed, but most of the parts were COTS or made with drills.

The modules shown in the picture were made with 100% Tetrix parts - don't do this if you can avoid it. We switched the side plates of the modules out for stronger plates and it vastly improved the performance and rigidity of the modules.

[Wren Hensgen]

Our CAD work, along with everything else the LANlords have created in Inventor over the years is available at GrabCad.

Link to the swerve modules here:

https://grabcad.com/library/lanlords-ftc-robot-design-2012-2013-1

[5383robotsrule]

thanks so much now i can convince my teammates to choose swerve drive

[Andrew Lobos]

thanks so much now i can convince my teammates to choose swerve drive

If you decide to go this route, I really recommend you try to build a prototype over the summer. We had a working testbed before the season started, and we were still improving and iterating on components during the competition season.

While FTC gives you a lot more freedom to make changes between events, implementing a complex system like swerve comes with a lot of challenges.

[5383robotsrule]

how did you guys drive the robot ?

[Andrew Lobos]

how did you guys drive the robot ?

The left gamepad stick controlled translation, and the right stick's X axis controlled rotation. Note that the left and right sides were chained together, so rotating and translating at the same time wasn't possible.