Alleviation of Viscoelastic Creep in Electrostatically Driven Soft Dielectric Elastomer Actuators Using Input Shaping Scheme

Dielectric elastomers (DEs) have significant potential in many applications, particularly in the realm of soft robotics, owing to their capability of undergoing large deformation. The viscoelastic creep response exhibited by dielectric elastomers when subjected to electrostatic input is an inherent characteristic of DEs that hinders their long-term performance and applicability. This behavior leads to time-dependent deformation during actuation, which limits their use as actuators in applications such as robotic grasping. This work presents control techniques that have been developed to minimize viscoelastic creep in dielectric elastomer actuators (DEAs). To simulate the viscoelastic creep behavior of DEAs under dynamic loading, a dynamic modeling framework is presented. The equations derived from the dynamic model are subsequently subjected to the two different control strategies using Linear Quadratic Regulator and Model Predictive Control. Control strategies have been devised to minimize creep and decrease the reaction time. The key aspect of both methods involve the development of an objective function, which is then optimized to determine the ideal control input voltage. The suggested control methods involve applying the fundamental optimal control concepts to an actuator. The actuator response on the application of a single-step input is observed. This principle is then extended to include the actuator response which is a multi-step signal using the proposed control strategy. The results suggest that the control methods are capable of efficiently dealing with the viscoelastic properties of the actuator to achieve different desired equilibrium conditions and hence optimize the performance of the actuator. The simulation results demonstrate that both the control strategies reduce the response time of the DEA by about 98% and also improve the steady-state response. This paper’s findings have the potential to be applied to the development of a control system for the mitigation of viscoelastic creep in DEAs.