Sometimes you see something on the internet and you just have to know more about the inception of the concept. For example, watch this cat ride a custom electric skateboard. We can probably all agree that it is awesome, but I can’t help but wonder why it was built, or how the heck you’d train a cat to even do that. I caught up with Kim Pimmel, an interaction designer for augmented reality at Adobe, and asked him about his project.
Why did you build this? What was your inspiration here?
I like to experiment at the fringes of UX design such as immersive interfaces in AR/VR, voice UX, or ML-driven design.
As a cat owner, I’ve been curious if I could create interfaces for my cat MIDI. My Pet Projects series investigates whether she can learn to use affordances such as buttons, levers and switches with a treat as a reward. So far I’ve made a robo-dog thing that gives out a treat when its nose is pulled and a machanical pinball machine – both of which she learned pretty quickly.
The latest project, an electric skateboard I built, has a throttle lever with a hole in it where treats go. Put a treat into the throttle lever, and she pulls it to get a snack and give the board a boost. It uses an Adafruit Feather microcontroller to sense is the throttle is pulled, and the Adafruit Feather DC motor featherwing to drive the motor on the board. It uses a high torque 2000rpm electric motor to move the board.
How did you build it? How long did it take?
The skateboard is a standard deck with chunky wheels for a smooth ride, and the drivetrain is an off the shelf electric skateboard belt system for longboards.
The throttle lever housing and electronics box were prototyped and built using a Glowforge to create the parts. The wooden throttle housing components were glued together with CA glue and baking soda for structural strength, sanded and painted to give it a finished look. The treat tray parts were cut from black acrylic since it doesn’t absorb food particles or kitty saliva, and can be easily cleaned.
A whole lot of prototyping went into landing on the final kitty skateboard design. Iterating fast = learning fast! pic.twitter.com/aJWSQiwG2R
— Kim Pimmel (@kpimmel) September 17, 2018
The project took several weeks to complete and went through multiple rounds of prototyping – from lo-fi proof of concept to polished final result. The very first version was a quick proof of concept made using a kleenex box and spare parts from the studio
That first POC used a capacitive touch sensor for the throttle, which worked great and was easy for her to use, but lacked the visual satisfaction of her pulling on a throttle to get the board moving.
What was the toughest part?
The hardest part of the build was finding the right motor that balanced torque and speed – it had to be strong enough to move the board and the kitty, but not so fast that she got spooked or crashed into things at high speed. I also didn’t want to spend too much on an out of the box electric skateboard motor system – that would also take the fun out of trying to make something custom. I tried several motors,finally settling on a brushed 2000rpm DC motor that the Adadruit DC motor featherwing could drive at both slower and faster speeds. I learned that DC motor control is a much more complex thing than I realized, and that easing into acceleration is key to a smooth off the line speed boost.
If you were going to build this again, what you do differently?
If I had to do something differently, I would have designed the electronics box so that it was easier to get into and tweak. It’s a bit difficult to get into, and has to be unscrewed from the board and flipped over to access the wires / battery / etc. Ideally there would just be a way to pop open that electronics box with a lid or access door.