Semi-Autonomous Control with Cyber-Pain for Artificial Muscles and Smart Structures

Abstract : Dielectric elastomer actuators (DEA) have characteristics that are the most similar to biological muscle, which is why they are also known as artificial muscles. The goal of this research was to establish the feasibility of using a cyber-pain signal to enable a DEA unit (comprising both an actuator and its local controller) to avoid some key failure modes. Our approach has built on our developments in dynamic self-sensing and realistic simulation of DEA electro-mechanics, including control circuits. The achievements of this project can be summarized as ? design of a test DEA that encapsulated the critical failure modes ? development of a novel monitoring circuit that incorporated our latest research into self-sensing algorithms ? measurement of leakage currents and discharges across the dielectric, both of which are indicators of the state or condition of the dielectric elastomer devices; and ? correlation of DEA quality with the frequency of partial discharges.