Apple Pi Offseason Swerve Bot

As a final project for our offseason training, Apple Pi created a swerve bot with a turreted telescoping arm on a wrist. The goal of this robot was to be able to track a Nalgene bottle with retro reflective tape on it, extend the arm, and bring to bottle in. The key components are as follows.

Full Robot - CAD - Photos

Swerve - CAD - Assembly Guide (people have reported issues with the CAD, I’ll update it tonight)

The newest and most significant part of the robot is our 2019 swerve modules. Two (all four in CAD) of the modules on the robot are based off of our new 2019 design. These modules were designed with one thing in mind: simplicity. The idea was to create a module that could be machined and assembled in the first week of build season, drastically reducing the traditional time sink of swerve. With this came significant reductions in complexity and weight. The most significant changes as compared to last years modules are:

  1. Box tubing for the lower assembly rather than 3 plates screwed together (inspired by 2767)
  • Eliminates screws, plates to be machined, holes to be drilled and tapped
  1. Driving the bevel gear off the motor shaft directly (inspired by 1533)
  • Eliminates a plate to machine
  1. Putting the bushing through the turning pulley rather than above it
  • Only one bushing rather than 2 need to be machined
  1. Belt driven (other than the bevel gear)
  • Reduces backlash
  1. Works with any 4in wheel up to 1in wide
  • Allows for use of treaded wheels
  1. VP drive encoders
  • What we have found to be the most reliable

The bushing and plates will be made on an Omio X6. The top hole pattern on the box tubing is made on the Omio then the bearing holes are drilled on the Bridgeport. The custom belt pulley will be 3D printed (most likely out of ABS).

The turret is a roughly 12" ‘lazy susan’ bearing. We made a custom plate to attach to the inner ring of the bearing and drive that using a belt.

Telescoping Arm and Claw
The telescoping arm is a modified version of 1/2 of our elevator for this year. We took one side off of our practice bot and modified it to work with a belt drive.
The claw is a prototype inner tube grabber from 2011 and not currently functional.

There is a lot more that could be said, and I am happy to, but in the interest of not creating an incredibly long post, please PM me or send me an email at: [email protected] (or ask questions below ) if you would like me to elaborate on anything.

The code for the swerve modules will be posted soon.

(Use the swerve specific GrabCAD post for the swerve CAD files as they are more up to date. The assembly guide is still very much a work in progress. GrabCAD previews didn’t seem to be working, so let me know if there are issues downloading the assemblies.)

1 Like

I’m impressed. Why did you decide to go with a turret? Were you not pleased with the precision or control you got from the swerve drive alone?

We thought it would be fun and a challenge are the main reasons. I doubt we would do something like this in competition season but who knows.

This a link to the code for the project. We program in LabVIEW :slight_smile:


I downloaded the CAD to look at it, but I’m wondering how you are securing the minicim+versa planetary down to the plate.


Owen has graduated so I’ll try to answer
They were attached using two #10 mounting screws through the plate to the VP. Normal set-up - could be they were not included in the CAD .

1 Like