3D Printed Electric Unicycle

Actually riding around at 30 km/h on a 3D printed means of transportation is pretty gnarly, if not foolhardy. So we were actually pleased when we dug deeper and discovered that [E-Mat]’s unicycle build is actually just a very nice cover and battery holder.

We say “just”, but a 3D-printed design takes a couple of cheap parts (the wheel and pedals) from the Far East and turns them into a very finished-looking finished product. Custom bits like this fulfill the 3D printing dream — nobody’s making it, so you make it yourself. And make it look pro.

It turns out that other people have noticed this motor/controller/pedal combo as well. Here’s some documentation to get you started.

It’s funny. Just four years ago, self-balancing powered unicycles were the realm of the insane hacker. Then came some hacker improvements, and now we’re at the point where you can mail-order all the parts and 3D print yourself a fancy enclosure.

Thanks [nonice] for the tip!

10 thoughts on “3D Printed Electric Unicycle

  1. I had attempted a Unicycle using a Brushless DC motor where I built the controller and self balancing logic using an mbed hardware. I was using PID controller as you have done so ( I believe), but got stuck in being able to drive at a constant speed. The issue was not with the motor drive, but rather with the control software that needs to work hard between maintaining the balance and providing a constant speed. The video on this link was what it was in 2011 and it didn’t improve beyond this. https://www.youtube.com/watch?v=6DlapRyK0c4

    Your video is very very cool and I wanted to shoot something like this. Can you throw some light on you controller software where the PID controller is good enough for maintaining the balance, but I had trouble with the outer loop. I tried using a speed loop for the outer loop on a smaller two wheel robot and it was fairly successful, but didn’t try on the bigger one since the unicycle is now in a dismantled state and the batteries are dead. Your insights would be greatly appreciated. My design didn’t graduate to using an LQR controller since I strongly believe that this problem can be solved with a double loop PID controller and I felt LQR controller was a bit complicated. ALso noticed that a complementary filter filter performed better than a kalman filter in IMU measurements (probably since I understood the complementary filter better and managed to tune ot well)


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s