Pi Zero Powered Skateboard

There’s something to be said for whizzing around town on your own automatic personal transport. It’s even better when you’ve built it yourself. That’s just what [The Raspberry Pi Guy] did – built a Wiimote controlled, Raspberry Pi Zero powered skateboard and whizzed around Cambridge to show it off.

It’s a fairly simple build – skateboard, battery, motor and mount, controller, Wiimote and Pi Zero. The Raspberry Pi controls the motor controller which in turn controls the motor speed. The Python code that [The Raspberry Pi Guy] wrote comes in at around a hundred lines and manages the motor controller and the Bluetooth connection to the Wiimote, which is used to control the board’s speed while the user controls the steering. [The Raspberry Pi Guy] says he’s gotten up to 30 km/h on the skateboard, which, given a powerful enough motor and a non-bumpy surface isn’t hard to believe.

It may seem a bit of overkill, running a bit of Python on a Raspberry Pi to run a motor (others have done it with something simpler) but it’s a fun project nonetheless. [The Raspberry Pi Guy] describes where he got the parts to put the skateboard together and has released the Python code on his GitHub page.

29 thoughts on “Pi Zero Powered Skateboard

    1. You know that in a lot of places PLCs running at only a couple MHz control io modules over modbus over very noisy lines with lots of errors and retries over 9200 baud? People complaint about not being “real time” but a raspi at 1GHz it’s a lot more “real time” than what most people think they have in prod. A raspi spending some cpu time on not work related tasks will delay the change in outputs a lot less than just the 9200 baud alone plus you can disable most of the system and get to a really bare bones linux with very little running in background.

      There are reasons to not use a linux computer as a PLC, many of which relate to reliability but no-one talks about that in these forums, all that people complaint is about real time, really get some “proper” Schneider PLCs (my experience) and measure the time it takes the system to react to an input it will be in ms (but not stable), do it again while using a io module and be ready to be surprised with it sometimes taking seconds to react. that’s way worse that what these hacks get on “muggle” PCs

  1. are there electrical skateboard out there that use some sort of foot control. Push you foot down to go faster? If your foot doesn’t push the throttle any more… it stops automatically. I can imagine all sorts reliability issues with a complicated processor board bouncing all day long, the trouble of booting, shutting down, connecting with the bluetooth remote. Paring problems connectivity lost… battery of remote is low (whaaa…. I can’t stop) and then seeing your build riding into the sunset (with you still lying on the ground looking at it). Anyway… I guess many people would have done this in a slightly different way. But then again… with a name “The Raspberry Pi Guy” how could he have used anything else. So I think we must realize that the build originated from the thought “what can I build using a pi” then “I want to build a electric skateboard and the only thing I could come up with was a pi”. And although the pi is a complex board by itself, it costs next to nothing and makes the use of a wii mote much easier then using a couple of 555 ICs to control the motor driver.
    So I guess I fully understand the reason for the pi and it is also a nice design challenge in software. I’m sure that he fixed all the problems I feared a few lines earlier. The build looks clean and well protected (although I hope the motor won’t get too dirty on a wet/dusty road. Anyway, I think this project has been a great success. I really liked the effort that was put into the video, which made it fun to watch.

    I can’t wait what this guy will come up next with, because with a name “The Raspberry Pi Guy” this is probably not his last pi project.

    1. we can reliably launch electronics at hundreds of g’s, accelerating and bouncing a bit on a normal road at a few miles an hour wont break any electronics that were mounted probably, solid state electronics are incredibly robust, the human riding the thing will break far faster.

      1. That is generally true, but not always, as in automotive ECUs with electronics like the one on the Pi (just CPU, power rails, connectors), if a fail can cost a human life, is not a naive problem to demonstrate that the component shall be dependable against the vibration. This lead to some type of capacitor selection w.r.t. other normally used, cover everything with some kind of epoxy, heatsinks are forbidden, and so on.
        I am saying this because some things often shall not be so obvious.

        1. you are right, that was what i meant by properly mounted, motor windings are potted for a reason.

          i wrote in response to the comment that vibration or the complexity of the processor in itself would be problematic, a more complex cpu often needs more complex support so more might be able to go wrong but with proper mounting that shouldn’t be a problem, discrete components give you more connections to mess up so they aren’t necessarily more reliable.

          1. I never stated it WOULD be problematic, I only wrote that I could imagine that it could be problematic. You are right that we can launch things with many G’s… but how many of those things are consumer electronics, like a the pi is? Please think really hard about that for a minute…
            I fully agree that simple through hole components also pose a risk and the more you have of them the higher the risk of a loose/broken contact, the bottom line is, the more complexity (no matter if it is mechanical/electrical or software) the higher the risk of problems. So KISS would be the first thing that would come to mind when I would be going to build something that would make me highly depending on it’s functionality. Not only regarding safety… walking home (or taking the bus) with a non functional piece of hardware can be as embarrassing as an silly accident.
            Then again…life is full of compromises.

          2. i was only making a comment on reliability and durability in relation to solid state technology, not critiquing the entirety of your post.
            consumer grade off the shelf components are exactly what many are launching or planning to launch in modern cubesats, there is a project called pi-sat, and several other projects are using them for secondary computation as well.

            again not saying that you aren’t right in adhering to the KISS philosophy only that complexity doesn’t necessarily equal lower reliability.

            some of the most complex systems on earth are also, when viewed as a whole, the most reliable.
            sure individual parts may fail but in a complex system where humans don’t have much of a chance of understanding everything there is an inherent need for abstraction to begin with, allowing for distributed failure is possible.

            no saturn V launch went without failed equipment as far as i know, but that was alright, it was expected and calculated into the operation of the machine.
            some complexity does bring more opportunities for something to go wrong and at some point it will be something one hasn’t planned for, nothing is ever certain, but that is true for simple systems as well

    1. Well, you aren’t edgy enough if your shirt isn’t buttoned up with a full row of Java-running iButtons and each of your heels isn’t running embedded Linux to count your steps… next up: color-changing LEDs running internally on Ubuntu Core…

  2. better idea, wireless wii nunchuck adapter with an attiny for processing. better yet, use a wired nunchuck. wii peripherals work on i2c so an attiny85 would work just fine. At least the buzzwords they used weren’t very expensive. Only thing funnier would be an intel edison to make Adafruit (R) NeoPixel (TM) running lights under the board blink with something something internet of things. That would hit all the hipster corporate buzzwords.

  3. I guess it is time to give up complaining about the “powered by” thing. Now, with headlines where vehicles that carry people are “powered by” CMOS gate transitions, I’m throwing in the towel. We will just have to use a unit of power when we really want to talk about power. “The new cars have more Wattness.” or something like that. More Jsec maybe?

Leave a Reply to iRubensCancel reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.