Robust Position Control of Electro-Hydrostatic Actuator Systems with Radial Basis Function Neural Networks

Abstract

This paper presents the robust tracking control of EHA (Electro Hydrostatic Actuator) system. EHA system provides the simple structure and frees from the environment pollution, but the poor tracking performance is caused by the internal leakage, surging of pump and friction. In order to increase the tracking performance of EHA system, ABSC (Adaptive Back-Stepping Control) with RBFNN (Radial Basis Function Neural Networks) is proposed. The proposed control scheme can guarantee the robustness and increase the tracking performance in accordance with the external disturbance and unmodeled uncertainties. The update law for the proposed ABSC with RBFNN is designed based on the reconstruction error. To verify the effectiveness of the proposed control scheme, computer simulation and experiments are executed and the results were compared with that of the general BSC(Back-Stepping Control).

Keywords

• electro-hydrostatic actuator
• friction
• adaptive back-stepping control
• radial basis function neural networks
• artificial intelligent