Custom Game Controllers/Pads

[davidthefat]

I have seen several teams in the past with custom made controllers or pads during competition. Would you say that there are advantages in having a custom game controller over a generic joy pad or game controller? Game controllers are made with video games in mind, that always does not translate to smooth robot driving and operating. There is a certain advantage in the fact that who ever makes the controllers will gain experience and have fun. Now, I personally still find controlling the arm with a joy pad to be clumsy. Would any of you say that having a custom game pad made for every game would make controlling the robot more intuitive?

Perhaps some unconventional ways to control the robot might win some awards, but I am talking about a more conventional style of controllers.

I am considering making a controller next year during the awkward first week for programmers where everything is in design phase and prototyping phase with not much programming at all.

[apalrd]

We have traditionally built a custom control box for our operator, and let the driver use either a Logitech DualAction or two joysticks.

This worked well until last year, with the introduction of the Cypress board. We used the Cypress board to handle IO in Kitty's Kat Box (Our operator that year was named "Kitty".). We had to boot up the Classmate at least three matches ahead of time in case the driver had an issue, which could force us to reboot one or two times. A very annoying fix. We needed an analog dial for kick power, and the Cypress board was the only thing that gave us an analog input that didn't re-center each time it was plugged in.

This year, we used an xBox controller (Kitty's Kat Pad) for the operator, but ran out of buttons (the disadvantage of using a stock gamepad). Unhappy with the Cypress board last year, we bought a Logitech Precision pad (which contains only buttons - no joysticks), opened it up, and soldered wires to all of the button inputs. We built a new box, which we backed in velcro so it sticks to the operator console shelf, and it included a 6-position auto select switch, 3 light color buttons, a button for Score LOW (never used in the official season), and an extra button (originally assigned to FOLD, it was just easier to fold it manually).

OTS gamepad:
+Already made
+Has JS axises
-Lacks many buttons (if you need a lot of buttons)

Semi-custom gamepad (using a gamepad as base instead of Cypress board):
-No support for analog inputs that don't automatically center
+Lots of buttons (Whatever the controller had, plus the 4 D-pad axis)

Cypress-based control box:
-Cypress bug - I don't know if this was fixed
+Easier to build than gamepad, maybe.
-Cypress bug. Seriously. Its a big issue.

As for HMI's, we've used a box of state buttons for many years with success. Our 2007 and 2005 robots both had a button for each state, plus buttons to open the claw/fingers, and 2007 also had ramp buttons. Our 2011 robot could have worked with such a control board, but we initially didn't want to use Cypress IO (and built around an xBox pad).

[msimon785]

Team 1515 has driven a fairly unique "gamepad" in the past 2 years- touch screen. We use a single 3 axis joystick to control the drive (both years mecanum) and the touch-screen to control everything else.
It's really quite cool. For instance, in Logomotion, we had 2 main screens: Drive and Maintenance modes.

Within Maintenance mode, there were animated thumbnails of the mecanum wheels, and FWD/Rev buttons, 2 speeds each. This allowed us to completely test the drivetrain. Addtionally, we had the ability to control both our forklift style lift and pneumatic joints in the arm. For the forklift, we had both absolute and relative control. We could also deploy the minibot.

Within Drive mode, every thing was streamlined in order to ensure speedy gameplay. To raise our lift, when on absolute control, we could touch any peg on the full onscreen pegboard, thus causing the lift to go to that height. (It was self-calibrating, every time it touched the bottom, it registered 0). When on relative control, we could simply press UP and DOWN arrows to take the lift to that height. Of course the pneumatics were simple, press and they change position. When the operator pressed the 'Deploy Minibot" button, an "are you sure" window popped up, and on OK, it deployed.

Consequently, we won the Innovation in Control in both the Los Angeles and Denver regional this year. The screen was a donation from a local company. The connection between the touchscreen and the classmate is somewhat complex, but it's worth it in my opinion. The image below displays our entire OI. I can't say I can explain how it was accomplished, so I'll end here.