Observer-based motion axis control for hydraulic actuation systems

Unknown dynamics including mismatched mechanical dynamics (i.e., parametric uncertainties, unmodeled friction and external disturbances) and matched actuator dynamics (i.e., pressure and flow characteristic uncertainties) broadly exist in hydraulic actuation systems (HASs), which can hinder the achievement of high-precision motion axis control. To surmount the practical issue, an observer-based control framework with a simple structure and low computation is developed for HASs. First, a simple observer is utilized to estimate mismatched and matched unknown dynamics for feedforward compensation. Then combining the backstepping design and adaptive control, an appropriate observer-based composite controller is provided, in which nonlinear feedback terms with updated gains are adopted to further improve the tracking accuracy. Moreover, a smooth nonlinear filter is introduced to shun the “explosion of complexity” and attenuate the impact of sensor noise on control performance. As a result, this synthesized controller is more suitable for practical use. Stability analysis uncovers that the developed controller assures the asymptotic convergence of the tracking error. The merits of the proposed approach are validated via comparative experiment results applied in an HAS with an inertial load as well.