Page 15 - Adaptive Identification and Control of Uncertain Systems with Nonsmooth Dynamics

4 Adaptive Identification and Control of Uncertain Systems with Non-smooth Dynamics

To achieve the proposed goal, we take inspiration from the adaptive
parameter estimation, adaptive control, and neural networks. Since these
studied non-smooth non-linearities have different characteristics, we will
consider each of them individually in various controller designs. Each con-
trol design is discussed in detail and provided gradually, with most chapters
building on their predecessors. It will be shown in each chapter how the
non-smooth dynamics can be modeled and adaptively compensated, and
how the desired control system response is achieved. The controller de-
signed by using adaptive technique and function approximation allows to
guarantee the system stability and to address the transient performance of
the tracking error.
Beyond the development of theoretical studies on the adaptive control
design and performance analysis, in order to better suit the requirements
of engineers, the book will also present application of the developed algo-
rithms based on servo systems, and showcase experimental results.

1.3 BOOK OUTLINE

This book includes 19 chapters, which are organized in five different parts.
Part 1 provides background information related to non-smooth charac-
teristics and states the motivation of this book. The overview of the book
and the preview of chapters are also provided.
Part 2 is concerned with the modeling and control for non-linear
uncertain systems with friction dynamics, which includes Chapter 1 to
Chapter 6.
In Chapter 1, the friction dynamics are discussed and several classical
friction models are reviewed. Moreover, two recently proposed piecewise
continuous friction models used in the control designs in this book are also
presented.
In Chapter 2, the parameter identification and control are investigated
for a servo system by using LuGre friction model. An intelligent glow-
worm swarm optimization (GSO) algorithm is used to identify the friction
parameters. Then, an adaptive sliding mode control is designed to achieve
output tracking.
In Chapter 3, an adaptive dynamic surface control (DSC) is presented
for speed tracking and torsional vibration suppression of two-inertia sys-
tems. The non-linear friction presented by LuGre model is combined with
echo state neural networks (ESNs). A prescribed performance function