Adaptive optimal tracking of a discrete-time minimum-phase plant under output uncertainty

We consider adaptive optimal tracking problems for a minimum-phase discrete-time plant with uncertainty in the output channel and bounded exogenous disturbance. The coefficients of the equation of the linear time-invariant nominal model, the uncertainty gain, and the upper bound of the exogenous disturbance are assumed to be unknown, and the worst-case asymptotic tracking error serves as the performance criterion. It is shown that the statement of the optimal problem depends on a priori information about the bounded control. The solution of the problems is based on the polyhedral estimation of all unknown parameters and the use of the performance factor of the tracking problem as an identification criterion for calculating the current optimal estimates. The performance and efficiency of the proposed control algorithms is illustrated by the results of numerical simulation.