Abstract
This paper is concerned with the design of an adaptive observer to estimate state variables and identify plant parameters of an unknown, SISO, linear discrete-time system with “fast modes” or “parasitic” elements, The observer design is based on the reduced plant model order equal to the order of the slow or dominant part of the unknown plant. The reduced-order observer strurture considered here is based on a non-minimal type representation with the adaptive law determined by equations that give an exponential rate of convergence. The performance of the adaptive observer is evaluated on systems with strongly and weakly observable parasitics and it is shown that the resulting performance is robust.
