Hand Force Estimation using Electromyography Signals

Abstract

In many studies and applications that include direct human involvement such as human-robot interaction, control of prosthetic arms, and human factor studies, hand force is needed for monitoring or control purposes. The use of inexpensive and easily portable active electromyogram (EMG) electrodes and position sensors would be advantageous in these applications compared to the use of force sensors which are often very expensive and require bulky frames. Among non model-based estimation methods, "Multilayer Perceptron" Artificial Neural Networks (MLPANN) have widely been used to estimate muscle force or joint torque of different anatomy of humans or animals. This paper investigates the use of Radial Basis Function (RBF) ANN and MLPANN for force estimation and compares the performance of the two methodologies for the same human anatomy, i.e. hand force estimation, under an ensemble of operational conditions. In this unified study, the EMG signal readings from upper arm muscles involved in elbow joint movement and sensed elbow angular position and velocity are utilized as inputs to ANNs. Moreover, the use of elbow angular acceleration signal as input for ANN is investigated. Towards this end, a single degree-of-freedom robotic experimental testbed has been constructed, which allows for data collection, training and validation.