Fuzzy supervised neurocontrol of electrohydraulic servos

The work presents theoretical and experimental investigations concerning the neuro-fuzzy control of electrohydraulic servos. The control algorithm is built around a neurocontroller designed to optimize a performance index. Whenever the neurocontrol saturates or a system performance parameter decreases, the control scheme switches to a feasible and reliable fuzzy logic control. In accordance with simulation studies developed in previous works of the authors, the neuro-fuzzy controlled system proves often to be better than the corresponding classical PI or LQG systems. We emphasize that the entire control synthesis is thought as a model free approach, thus ensuring reduced design complexity and focusing on increased system robustness. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)