Modeling and adaptive pseudo-inverse control for dielectric elastomer actuated motion control systems

In this paper, we propose a neural network approximator-based proportional-derivative pseudo-inverse control scheme. The purpose is to precisely control a motion control platform actuated by a dielectric elastomer actuator (DEA). Our main contributions are as follows: (1) a new butterfly asymmetric shift Prandtl-Ishlinskii (BASPI) model that can describe the butterfly hysteresis behavior in a DEA; (2) the butterfly hysteresis pseudo-inverse compensation algorithm to effectively mitigate the butterfly hysteresis, instead of the explicit butterfly hysteresis inverse compensator. The algorithm searches for the practical control signal from the hysteresis temporary controller; (3) a DEA motion control platform is constructed. Finally, we conducted the open-loop and closed-loop experiments to verify the effectiveness of the proposed BASPI model and the proposed control scheme.