Global adaptive output feedback controllers with application to non-linear friction compensation

In this paper, we consider a class of non-linear systems in which a set of constant parameters is unknown and some state variables are not available for measurement. For such systems we provide a constructive procedure for the solution of the global adaptive tracking problem with dynamic partial state feedback. We illustrate an application of the control strategy to the adaptive non-linear friction compensation of a DC motor servomechanism. We improve previous results in tow directions: we allow for a subset of the unmeasurable states to enter in a system non-linearly; we consider systems which are linearly parametrized with respect to a set of unknown constant parameters.