Title :
Non-asymptotically stable observers for linear time-invariant systems
Author_Institution :
Universit?? de Montr??al, Montr??al, Qu??., Canada
Abstract :
The concept of asymptotical observation is extended to include a new kind of observers which converge asymptotically not to the (unknown) state of a given system as conventional observers do but rather to an arbitrarily small neighborhood of the state. This type of convergence represents a natural technical requirement in applications and leads to the broadest class of models for asymptotic state estimation. Non-asymptotically stable observers are shown to be robust and to possess closed-loop stability properties under permanently acting disturbances. They require a bit of additional pointwise information and detectability condition is no longer necessary. In linear time-invariant case such models are conditionally stable in the large in the sense specified below.
Keywords :
Asymptotic stability; Control design; Convergence; Eigenvalues and eigenfunctions; Equations; Observers; Real time systems; Robust stability; State estimation; Yield estimation;
Conference_Titel :
Decision and Control including the 16th Symposium on Adaptive Processes and A Special Symposium on Fuzzy Set Theory and Applications, 1977 IEEE Conference on
Conference_Location :
New Orleans, LA, USA
DOI :
10.1109/CDC.1977.271612