Tuning of EMG sensor gain
The completed exo arm
Best Hardware Hack
Just a year ago all of us were serving our national service. We got in touch with a world in which people have to perform heavy lifting on a regular basis. Such heavy lifting is tiring and often is very strenuous on emergency response personnel, armed forces, nursing and construction work. In addition to doing heavy work, the exoarm can also assist stroke patients with their rehabilitation. The stroke patients benefit from consistently practicing the usage of their muscles. We decided to draw inspiration from fictional characters like Iron Man and movies like Elysium to design a solution that helps enhance the strength.
What it does
We have used electromyography to detect muscle motion and design an exoarm that follows the wearers commands and (somewhat?) enhances their strength.
How we built it
We used an old arm brace along with a high torque motor and fishing line for actuation. We also included EMG sensors to detect motor neuron signals. These signals were processed on board the Arduino which then controlled a motor to pull the fishing line.
Challenges we ran into
There were numerous challenges. Firstly the mechanical part of the project was particularly challenging. As most of us are from a software background we found our initial design to be inefficient. In fact lifting a load like a human arm is not an easy task. We eventually reworked our pulley system to make it possible to lift loads.
Detecting EMG signals was also a challenge as the gain of the amplifier needed to be tuned to be perfect. There was a fair deal of noise and furthermore, electrostatic build up in the human body made it hard to detect EMG signals. We also found that the tricep signals were often overwhelmed by bicep signals. We developed a very simple algorithm to overcome these challenges.
Accomplishments that we're proud of
Making an exosuit that actually can lift things is something that was pretty awesome. It made us feel like Tony Stark.
What we learned
Mechanical design is very hard. So is detecting EMG. Actually designing something like this requires a lot of trial and error.
What's next for EY - ExoArm
Make it into a full suit. PLS GIVE US $$$$....