Robust tracking and disturbance rejection — nondynamic multirate output feedback and observer-based control

A new nondynamic multirate output feedback approach to robust tracking and disturbance rejection is presented. Assuming that a linear model of the plant under nominal operating conditions and under failure conditions is known, convex programming can be employed to find a state feedback gain such that the closed loop system meets given specifications in all operating points. If the full state is not available for feedback, then under a certain assumption, a multirate output feedback controller can be constructed which realizes the effect of this state feedback gain in all operating points. By contrast, if an observer is used, an estimation error under failure conditions leads to a deterioration of performance. This point is illustrated by simulation and experimental results with a laboratory aircraft model.