This monograph bridges the gap between the nonlinear predictor as a
concept and as a practical tool, presenting a complete theory of the
application of predictor feedback to time-invariant, uncertain systems
with constant input delays and/or measurement delays. It supplies
several methods for generating the necessary real-time solutions to the
systems' nonlinear differential equations, which the authors refer to as
approximate predictors.
Predictor feedback for linear time-invariant (LTI) systems is presented
in Part I to provide a solid foundation on the necessary concepts, as
LTI systems pose fewer technical difficulties than nonlinear systems.
Part II extends all of the concepts to nonlinear time-invariant systems.
Finally, Part III explores extensions of predictor feedback to systems
described by integral delay equations and to discrete-time systems.
The book's core is the design of control and observer algorithms with
which global stabilization, guaranteed in the previous literature with
idealized (but non-implementable) predictors, is preserved with
approximate predictors developed in the book.
An applications-driven engineer will find a large number of explicit
formulae, which are given throughout the book to assist in the
application of the theory to a variety of control problems. A
mathematician will find sophisticated new proof techniques, which are
developed for the purpose of providing global stability guarantees for
the nonlinear infinite-dimensional delay system under feedback laws
employing practically implementable approximate predictors.
Researchers working on global stabilization problems for time-delay
systems will find this monograph to be a helpful summary of the state of
the art, while graduate students in the broad field of systems and
control will advance their skills in nonlinear control design and the
analysis of nonlinear delay systems.
