A study of real-time EMG-driven arm wrestling robot

An EMG-driven arm wrestling robot (AWR) is being developed in our laboratories for the purposes of studying neuromuscular control of arm movements. The AWR arm have 2-DOF, integrated with mechanical arm, elbow/wrist force sensors, servo motor, encoder, 3-D MEMS accelerometer, and USB camera, is used to estimate tension developed by individual muscles based on recorded electromyograms (EMGs). The surface electromyographic signal form the upper limb is sampled from a real player in same conditions. By using the method of wavelet packet transformation (WPT) and auto regressive model (AR), the characteristics of EMG signals can be extracted. Artificial neural network is adopted to estimate the elbow joint torque. The effectiveness of the humanoid algorithm using torque control estimated via WRT and neural network is confirmed by experiments. The purpose of this paper is to describe the design objectives, fundamental components and implementation of our real-time, EMG-driven AWR arm.