Adaptive position control scheme with neural networks for electro-hydrostatic actuator systems

This paper deals with the robust position control of electro hydrostatic actuator(EHA). In general, the position control of EHA systems based on the model of itself is difficult because of system uncertainties such as parameter perturbation, friction, and external disturbance. To solve the problems due to these system uncertainties, an adaptive back-stepping control (ABSC) scheme with radial basis function neural networks (RBFNN) is proposed. The adaptive back-stepping controller consists of back-stepping controller and adaptive rule for reconstruction error. Moreover, to estimate the bounded uncertainties of the reconstruction error, the RBFNN with online update law is designed. The effectiveness of the adaptive back-stepping control system with RBFNN was compared with that of the standard back-stepping control system through computer simulation.