Page 9 - Adaptive Identification and Control of Uncertain Systems with Nonsmooth Dynamics

PREFACE

Human ambition in exploring natural phenomena and manufacturing in-
telligent products has continuously created emerging methodologies and
technologies, among which the modeling, identification, and control
(MIC) have been proved as a multidisciplinary subjective to tackle engi-
neering problems. In particular, with the rapid development of information
science and intelligent manufacturing techniques, many industries have un-
dergone great changes, where the production equipment and industrial
processes are becoming more complex, such as sustainable energy, manu-
facturing, robotics, mining and metallurgy, etc. For these complex systems,
non-smooth non-linearities, e.g., friction, dead-zone, saturation, and hys-
teresis, are usually unavoidable due to the use of mechanical components,
hydraulic and piezoelectric actuators, power transmission, and other elec-
tromechanical devices. Such non-linearities are usually difficult to model
since they may vary with time and have non-smooth dynamics. They
could severely deteriorate the control system performance and even trig-
ger instability. In fact, the existence of such non-smooth dynamics leads to
significant challenges in the control designs. Hence, modeling and control
of systems with non-smooth non-linearities have always been an important
and active research area in the control field.
Duringthepastfew decades, innumerableefforthasbeen madebythe
research community towards the modeling and control of systems with such
non-smooth non-linearities, and great progress has been achieved in re-
cent years. In brief, two different control methods have been developed
to compensate for the effect of such non-smooth dynamics, e.g., inverse
model based control, and intelligent control. However, there are still cer-
tain constraints imposed in these available approaches, which restrict their
practical applications. For instance, the inverse model based compensation
methods presume that accurate models of such non-smooth non-linearities
should be available; the non-smooth characteristics also make the analy-
sis and online tuning of intelligent control difficult. Hence, it still remains
demanding, yet challenging, to develop new techniques suitable for model-
ing and control for systems with non-smooth characteristics and unknown
non-linearities, and to explore their applications in practice. This is more
xvii