Flexi Bird

McDonalds
SinA
Orko & Orko
Jayden Bang

Inspiration

Electromytography has been a medical diagnostic tool that has fascinated me for years. Personally, having been a patient of heart block, I semi-fondly remember when I'd get electrodes stuck to me and an EKG machine left in my pocket. I was clueless as to how this device functioned. Now, years later, we got the unique opportunity to understand how this sensor works and build an application around it.

What it does

Our goal was to use the potential difference from muscle contractions and be able to monitor the potential difference as well as the ability to use this voltage as a form of controller in a game. We wanted to encourage "getting built" so we made the emphasis of our project to encourage bicep curls.

How I built it

We used the Myoware sensor and an Arduino Uno to collect this data. We then created our own Flexi Bird game in Python using the pygame library that we could control by flexing. To collect data and analyze it, we used a combination of matplot and numpy.

Challenges I ran into

The biggest problem is that none of us are particularly strong or 'built'. We lack endurance. To counteract the problem of getting tired we created an autoregressive moving average model that looks at past data to produce a new threshold. By using a game, we can define a win or loss and also missed possibilities. The game starts at a higher threshold value that gradually decreases with an increased number of losses. This ensures both safety, by lowering the required flex to flap (F2F) as time goes on preventing strain, and also a complete workout. It also means that we can work out any part of our body by attaching the electrodes.

Also none of us really brought any exercise equipment, so we kind of just used bottles. A lot of bottles filled with water.

We also experimented with an IoT sensor, but that ended up killing my computer (it probably wasn't the microcontroller), which was very unfortunate. I cried a lot. We abandoned it :(

Accomplishments that I'm proud of

This was the first kind of hardware project that we had ever made, which was quite interesting. It was really neat seeing how our own interactions were translated to a computer using the Myoware sensor. I also liked our thinking with the filling-boxes-with-water-bottles-filled-with-water to act as rudimentary exercise equipment.

What I learned

How to create an autoregressive model
How to use an Arduino Uno
How Electromytography worked.
How to utilize signal processing and cleaning data to make it readable for our python code and