Repositioning control of a two-link flexible arm by learning (Q1361311): Difference between revisions

This paper considers the repositioning control of a two-link arm, the second one being flexible, using a learning strategy. It is shown that by using state feedback the nonlinear system can be transformed into a perturbed linear one, for which a general theory has been presented previously. It is proved that the proposed learning algorithm is robust, not only with respect to sufficiently small nonlinear plant perturbations, but also with respect to unmodeled high-frequency dynamics. The experimental results, presented at the end of the paper, confirm the validity of the approach.