EMG-based hand gesture control system for robotics

Abstract

In this paper, a Electromyogram (EMG) based hand gesture control system is developed. A wearable human machine interface (HMI) device is designed for an in-home assistance service robot. An EMG-based control system utilizes MyoWave muscle sensor to acquire and amplify EMG signal. A microcontroller system is used to an artificial neural network (ANN) to classify the EMG signal. Based on different hand movements, commands are sent through WiFi to control the motor in a service robot. The on-board Camera system mounted the robot can capture video real-time. In addition, a web server is implemented to provide live video feedback for robot navigation and user instructions.