Adaptive H/sub /spl infin// control using coprime factors and set-membership identification with application to a missile autopilot

A methodology for the synthesis of adaptive controllers is developed, which combines the H/sub /spl infin// control, mixed /spl mu///spl nu/ analysis and robust estimation theories. On the one hand, the discrete-time control law is designed using McFarlane and Clover's procedure. On the other hand, set-membership techniques are applied to the estimation of a state-space model in the presence of neglected dynamics and measurement noise. As the set-membership estimator provides a measure of the parametric uncertainty in the estimated model, it becomes possible to bypass the certainty equivalence principle and adjust online the achievable performance of the control law with the parametric uncertainty. Stability and robustness properties of the adaptive scheme are proved using the finite-time convergence of the estimates. The methodology is applied to the synthesis of an adaptive H/sub /spl infin// missile autopilot.