Natural Robot Control Interfaces Using Wearable Sensors

The JPL BioSleeve is a gesture control interface that is worn on the forearm, which contains an array of surface electromyogram sensors to read electrical activity created by muscle flexion and extension, and inertial sensors to measure arm motions. An embedded processor then uses machine learning algorithms to recognize activity patterns and generate control commands from hand gestures. The goal of this project is to use Matlab and Unreal Engine to convert the muscle activity to inputs used to control a simulated robot in a virtual environment. To prove viability we are creating a virtual environment with a model of the International Space Station (ISS) to navigate a SPHERES robot in an Unreal simulation. There are various other means of controlling robots, but many are cumbersome and limit the user’s ability to interact with the objects in their vicinity. By developing a wearable gesture control interface, the person controlling the robot will be able to interact with the environment and control the robot in a more intuitive nature