This project will allow people with minimal or zero fine motor control interact with devices that would ordinarily require more controlled hand or arm gestures. By simulating motions such as swipe right, swipe left, push down, or pull up, the program can interpret very inconsistent or shaky movements as a single command. Because the movements do not require force, the program can also be used to simulate button presses for those who otherwise lack the arm strength to activate buttons which require pressure.
Moving forward, there are two areas of research and improvement that would be required before this project would be testable in a real world situation. The first is the interaction between the reader program operating on the host device and any API’s on that device that would be required to translate input commands to generic control functions. For example, translating the swipe right or swipe left to a forward or backward in a PowerPoint presentation. To make this project fully unctional, we would also need to alter the program to operate at a much faster sample rate, and with more sensors. The current limitations of the default teensy setup limit us to four samples per second. But with more optimized hardware and software, our design and setup is capable of getting close to twenty sets of data reads per second from four different sensors. Once we are able to implement those optimizations, we believe that this system is very capable of interpreting spastic gestures with a high degree of accuracy.