A Low Cost sEMG Development Platform for Hand Joint Angle Acquisition

Abstract

A consolidation of sEMG to Muscle Force signal processing and Fingertip Workspace Mathematics (FWM) is hypothesized in this literature. Consequently, this hypothesis suggests a projection matrix from muscle force to joint angles of the hand. Using a supervised kinematic algorithm, an sEMG device can learn to describe an individual's fingertip positions in two steps. The first step is inverse kinematics to learn a projection from joint angle to muscle force. The second step is forward kinematics of muscle forces to predict joint angles without direct observation. This literature presents low cost hardware design for acquiring forearm sEMG signals and fingertip joint angles. The consolidation of sEMG to muscle force and kinematic hand modeling bridges the gap between physiologic research and human interfacing technology.