Output regulation of flexible mechanisms by learning

The problem of regulating the output of a flexible mechanism by means of learning algorithms is discussed. A general methodology is presented, and its applicability in the face of neglected high-frequency dynamics is investigated. The learning algorithm introduced is of the gradient type and its convergence has been at least proven in the case of sufficiently small elastic displacements. The performance of the proposed learning algorithm is illustrated by means of an academic example.<<ETX>>