Actuation of an Ionic Polymer Metal Composite Actuator with EMG Signal by Using Fuzzy Clustering Method

Ionic polymer metal composites, IPMCs, are a novel class of electro-active polymers (EAPs) which bends in response to a relatively low electrical voltage due to the motion of cations in the polymer membrane. IPMC materials have wide range of applications in robotics, biomedical devices and artificial muscles. This paper presents a fuzzy logic approach to electromyogram (EMG) pattern recognition for an IPMC actuating system with EMG signal. EMG signals generated by the contraction of muscles in the human forearm were used as an electrical stimulus for actuating the IPMC actuator. EMG is a method of recording and quantifying the electrical activity produced by muscle fibers of motion units. Fuzzy inference system can be used to map an input feature onto an output class. Fuzzy data clustering was used to categorize the muscle signals and recognize the contraction of the muscle. Also we need to consider the mechanical design matters such as light weight and small size with flexible behavior. The IPMC in particular has been vastly applied to the artificial muscle because it is driven by a relatively low input voltage. This EMG signal generated from the human flexor Carpi ulnaris muscle was pre-amplified before transferring to IPMC for achieving the large bending behavior of this actuator. The experimental results, confirm ability of IPMC as artificial muscle which actuating with EMG signals.