The hidden science and art underlying skateboard tricks is a source of fascination. This device 3 dimensionally tracks the skateboard position during the execution of tricks. The data and trails of the tricks are then used to create sculptures and immortalize the motion of the skateboard and the style of the individual skateboarder. Amongst the possible applications for this project, my work focused on creating artistic visualizations from skateboard tricks.

There were two main technical challenges with this project:

Achieve spatial position tracking with small electronics and without cameras and computer vision.
Create a sturdy and compact case for the electronics

Step 1: Electronics

This tutorial is using the following electronics:

Adafruit feather 32u4 Adalogger (Arduino based board great to directly log data to a sd card)+ micro sd card
VL53L0X Time of Flight Distance Sensor (we use to measure distance from ground to skateboard when possible)
Adafruit BNO055 Absolute Orientation Sensor (to get the skateboard orientation)
Sunfounder Hall Sensor + magnets (this is used for getting the skateboard speed)
Red LED + resistors
Push button + resistors
Small battery

Follow the schematics for wiring if you don't want to change the arduino code.

The arduino application to record the data is available at this link:

https://github.com/paul-ferragut/skateboardTracker

Step 2: Truck Case

The main difficulty for this project is to build a case able to resist the tricks impact. You can try 3D print a case however it might break with the first tricks...

I was lucky enough to have access to a Haas Mill lathe able to machine aluminum. After iterating a 3D printed prototype I milled a final version in aluminum. I decided to only build the base plate part of the truck and re-use the top part and the bushings.

The 3D model of the truck is available at this link:

http://a360.co/2sZOmg8

The part was tested to fit with this independent truck model.

If you use the same base plate model you will need to pack the electronics very tightly in the case.

Step 3: Wheel Odometer

I find it useful to get the speed data of the skateboard. In order to get this data, I used a Hall sensor and a magnet on the wheel. The hall sensor detects every wheel revolution thus it becomes possible to calculate the speed. You will need to put a magnet on the wheel. Be careful to get the orientation of the magnet right in the wheel, it will work only in one direction, to find the magnet direction you can plug the hall sensor and test.

The hall sensor is put in a 3D printed case and attached to the wheel.

Wheel 3D model: http://a360.co/2t9y4fD

Step 4: Go Skate!

Step 5: Software

Arduino recordings contain the following data:

//New record:2396<br>qW:-0.6567;qX:0.1840;qY:-0.6637;qZ:-0.3072;hall:0;altitude:8190;time:330909qW:-0.6548;qX:0.1886;qY:-0.6617;qZ:-0.3127;hall:0;altitude:8190;time:330936qW:-0.6545;qX:0.1901;qY:-0.6603;qZ:-0.3153;hall:0;altitude:8191;time:330981qW:-0.6554;qX:0.1937;qY:-0.6545;qZ:-0.3234;hall:0;altitude:8191;time:331024qW:-0.6553;qX:0.1956;qY:-0.6503;qZ:-0.3309;hall:0;altitude:8191;time:331069qW:-0.6534;qX:0.1962;qY:-0.6470;qZ:-0.3403;hall:0;altitude:8191;time:331112qW:-0.6521;qX:0.1956;qY:-0.6440;qZ:-0.3488;hall:0;altitude:8191;time:331155qW:-0.6494;qX:0.1931;qY:-0.6407;qZ:-0.3611;hall:0;altitude:8190;time:331202qW:-0.6425;qX:0.1929;qY:-0.6429;qZ:-0.3698;hall:0;altitude:8191;time:331245

eg this flip back

Interpreting the data is a challenge, I gave it a try with this bespoke openFrameworks application:

https://github.com/paul-ferragut/skateTracker

You might be interested in using the data with other software of your choice.

In this software, I manually define the beginning and the end of the trick in the air. I believe someone more advanced in mathematics might have suggestions to improve the software, I am keen to hear suggestions.

The trails are smoothed and I added features to create particle effects around the trail.

Step 6: 3D Prints

36flip by PaulFerragut on Sketchfab

The software is using a library to convert the trail into meshes in order to export to 3D models...

in this fork from Neil Mendoza ofxPtf there is a function to export pipes into meshes: https://github.com/paul-ferragut/ofxPtf

I used a Fortus 450mc to print wheel trails, additionally, I tested a ZPrinter 3D printer for more detailed color prints.

Step 7: Next Steps

This project was great fun, especially working at Pier 9 in San Francisco which is next to many famous skate spots. In the future, I would like to improve the software and test other electronics as well 3D print larger tricks sculptures.