Participation in the development of a prototype of an opensource hand prosthesis.
This project took place in Medialab Prado in Madrid, with the research group Autofabricantes.Medialab is a center that hosts different types of projects, ranging from arts to science. Autofabricantes is a group that seeks theoretical and technical advancements in the self fabrication of human prosthesis. This project stablishes an ethical and political alternative to the current system of patents and overcosts in the closed industry of human prosthetics.
During my collaboration in this project, I worked with the group of electronics. Together we developed the state machine that would determine the ways to control the hand, given the information that we could obtain.
We also worked taking design decisions concerning the hardware and electronics. During this particular time, we developed a hand for a kid born single handed. She had both wrists working, therefore we focused in the development of the motion of the fingers of a 3D printed hand.
As a datasource, we used a Myoware Mio Sensor. When we move a muscle, it generates electric potential. This sensor measures this electric potential and processes the electrical activity converting it into a simple analog signal that can easily be read by Arduino. The gif attached belongs to the kikkstarter campaign of the sensors, and I think it provides a good understanding of the method.
To generate the movement of the fingers, we used 3 servo motors, that would generate the tension for contracting the fingers.
As a datasource, we used a Myoware Mio Sensor. When we move a muscle, it generates electric potential. This sensor measures this electric potential and processes the electrical activity converting it into a simple analog signal that can easily be read by Arduino.
To generate the movement of the fingers, we used 3 servo motors, that would contract the fingers. As a data source, we were controlling two signals coming from two sensors.
These are two possible state machines we thought about when considering the interaction between the kid and the prosthesis. We observed a difficulty in keeping the incoming signal from the muscles constant.This signal was even more difficult to track when the measured muscles are not developed naturally. Therefore, we programmed two different state machines, where the time off, together with the thresholds for the received signals were personalized for the user's comfort.
In order to find the best muscles to track, as well as finding natural interaction, we worked together with physiotherapists to position the sensors. Finally, for this first prototype we located each sensor on the arm and forearm.
Medialab Prado, Autofabricantes