Adaptive Inverse Compensation Output Feedback Control for Dielectric Elastomer Actuator Motion Control System

In this paper, an adaptive inverse compensation output feedback control scheme is designed for the dielectric elastomer actuator (DEA) motion control system. First, the controller adopts the radial basis function neural network (RBFNN) approximator, which avoids the parameter identification of DEA motion control system. Second, a long short term memory neural network-based hysteresis inverse model (LSTMNN-BIM) is trained to compensate the hysteresis nonlinearity. Third, by designing a high-gain state observer to estimate the unknown state variables of the system, only the output of the system needs to be measured. Stability analysis shows that all signals in system are eventually uniformly bounded. Finally, simulation experiment is conducted, and the experimental results demonstrate the effectiveness of the proposed control scheme.