TAMU Computer Science: Semgo (video)

The goal of this project was to develop a physiological interface that allowed a user to remotely control a mobile robot with muscle movements. The students designed an embedded system that processed signals from four electromyogram (EMG) sensors using frequency analysis. Once the EMG activity had been classified into one of five different actions (left, right, forward, backward, stop), the corresponding command was transmitted to a mobile robot through an RF transmitter.