Ada Developers Academy Capstone Project
Goal: Create an application that will record the motion data from my horse moving and identify if she was walking, trotting or cantering.
After programming for 6 months, capstone time rolled around at Ada. I had this idea of a device that would track workouts on horseback - kind of like a horse fitbit.
The ‘first generation’ hardware solution was a raspberry pi and accelerometer/gyroscope. A fun solution, but not ‘user friendly’ as you can see it taped to my saddle here:
I quickly moved onto version 2.0 - my phone. An easy solution using the android accelerometer function. I wrote a front end to capture the data points, then load them to the backend for the algorithm to process. The app was written in Node.js and Javascript.
I used Python and a machine learning library called scikit learn to process the data. After training a k-Nearest Neighbor algorithm I could quickly and accurately categorize thousands of data points.
After the algorithm had done its work I could display time spent in each gait to the client.
This project was an amazing, immersive experience, truly the icing on the cake that was my time in the classroom at Ada Developers Academy. I can’t believe in such a short time I learned the skills to take on a project like this and look forward to expanding my skills even more during my internship at Samsung SDSA on the Cloud Native Computing Team.