Robust Control of Elastic Robots

In this thesis the performance of industrial robots is improved by modern control theory. A new method is presented in order to identify LPV (Linear Parameter Varying) models with low complexity. Furthermore, existing controller synthesis algorithms are modified and developed. These algorithms are then applied based on the identified models. Unmodelled dynamics and uncertainties are considered in these syntheses in order to guarantee robustness of the closed loop system. Finally, the controllers are validated on the hardware in the loop testbed.