LAUSR

In the paper the twofold problem of stabilization and output adaptive tracking is addressed for the class of multi-input multi-output (MIMO) linear time-invariant (LTI) unstable plants with known parameters, unmeasurable state, and known distinct input delays. The reference is represented by the vector of multi-harmonic time functions and is generated by an autonomous linear dynamic model (exosystem) with known order but unknown parameters. The amplitudes, phase shifts, and frequencies of the harmonics of these functions are unknown. The solution proposed is based on a robust predictorfeedback stabilizing control law, suitable parameterization of the tracking error, special implementation of the augmented error scheme, and direct adaptation algorithm providing asymptotic tracking without identification of the exosystem parameters. The stabilizing part of control palliates negative influence of distinct input delays. Regardless of the values of input delays, the adaptive control law designed ensures boundedness of all signals in the closed-loop system and drives the tracking error to zero.