On the use of robust regulators in adaptive control

Using concepts from non-adaptive robust control, the stability and robustness of a general indirect adaptive control algorithm is analyzed. The true plant is assumed to have bounded perturbations on its normalized coprime factors; allowing inclusion of a larger class of perturbations than is normally considered. It is shown that the use of robust controllers, particularly those obtained from H^∞ optimization techniques, can improve local robustness as long as the controller parameters are sufficiently smooth functions of the estimated parameters. Also shown is that a controller designed to minimize the effect of plant perturbations will also be robust to perturbations to the controller itself.