HIGH-PERFORMANCE ROBOTICS WITH SOFT ACTIVE SYSTEMS

Abstract

DEA is a promising candidate for robotic applications, primarily due to its large actuation strains and high specific energy. Their performance can be enhanced by conceptual design or by material development. We provide an analytical framework to identify optimal conditions for a very large continuously electrically-induced actuation (>500%) of an acrylic-based DEA. An ‘antagonistic cycle’ is then developed with inspiration from the agonist-antagonist pairs of human muscle. This design facilitates higher specific energy outputs and easy integration into robotic systems. An analytical model is performed to reveal optimal conditions that realizes an antagonistic system with large actuation and 75 mJ/g work output. A semi-empirical method is used for practical DEAs to build a continuously tunable robotic arm. Finally, a molecular-based constitutive model is adopted to predict DEA responses and reveals new instability mechanisms. We hope that the findings in this dissertation will broaden the field of application for DEAs.