Parameter identification for improved performance of model-based control of hydraulic manipulators

Abstract

In addition to hydraulic manipulators’ complex and nonlinear dynamics, their dynamic behaviour typically also involve significant parameter uncertainties. This study aims to evaluate the performance of nonlinear model-based control of hydraulic manipulators, by comparing (i) a controller using model parameters acquired from manufacturer datasheets and computer-aided modeling tools, in relation to (ii) a controller exploiting parameter identification methods to estimate the model parameters from pressure transmitter and joint encoder data collected from the plant. Experimental results show that the proposed control scheme relying on the identified parameters results in significant improvement of Cartesian position tracking performance in free-space motion, in comparison to using the nominal parameters.