Finite-Time Stability and Stabilization of LPV Systems
HU Yanmei
1
, DUAN Guangren
1
, TAN Feng
1
1. Harbin Institute of Technology, Harbin 150001, P. R. China
E-mail: g.r.duan@hit.edu.cn
Abstract: This paper deals with the finite-time stability analysis and design problems for Linear Parameter Varying (LPV)
systems. First, we extend the concept of the finite-time stability to LPV systems. Then, sufficient conditions for finite-time
stability are presented by using Lyapunov-like functions. Moreover, based on the previous finite-time stability analysis, we
explore to find the sufficient conditions of the finite-time stabilization problem via designing parameter-dependent state feedback
controllers. Further, these synthesis conditions are formulated in terms of Linear Matrix Inequalities (LMIs) that can be solved
via efficient interior-point algorithms. Simulation studies illustrate the effectiveness of the proposed approach.
Key Words: Finite-time Stability, Finite-time Stabilization, Linear Parameter Varying Systems, Linear Matrix Inequalities
1 Introduction
Practical systems often have nonstationary or nonlinear
behavior, and their dynamics rely on external variables [1].
The LPV systems provide an attractive modeling framework
which has the ability to describe special classes of non-linear
and/or time-varying phenomena. Due to the characteristics
of this class of systems, LPV systems have been attracting
considerable attention over the past few decades. For recent
results, readers can see references ([2–9] and the references
therein).
The concept of finite-time stability dates back to the
1950s, when it is introduced in the Russian literature [10–
12], later during the sixties this concept appeared in the west-
ern control literature [13, 14]. Roughly speaking, a system
is said to be finite-time stable if, given a bound on the initial
condition, its state does not exceed a certain threshold dur-
ing a specified time interval. It is mentioned that finite-time
stability and Lyapunov asymptotic stability are independent
concepts; indeed a system can be finite-time stable but not
Lyapunov asymptotically stable, and vice versa. While Lya-
punov asymptotic stability deals with the behavior of a sys-
tem within a sufficiently long (in principle infinite) time in-
terval, finite-time stability is a more practical concept, useful
to study the behavior of the system within a finite (possibly
short) interval, and therefore it finds application whenever
it is desired that the state variables do not exceed a given
threshold (for example to avoid saturations or the excitation
of nonlinear dynamics) during the transients. Some works
dealing with analysis and control design of finite-time con-
trol problems have been recently published, see for example
references [15–20]. However, there exist few results for the
finite-time stabilization problem of LPV systems.
Motivated by the above-mentioned observations, the
finite-time stability analysis and design problems for LPV
systems with unknown parameter variation rates are consid-
ered in this paper. The definition of finite-time stability is ex-
tended to LPV systems. By using Lyapunov-like functions
This work is supported by the National Basic Research Program
of China under Grant No.2012CB821205, by the Innovative Team
Program of the National Natural Science Foundation of China under
Grant No. 61321062 and 61503100, and also by Self-Planned Task
NO.SKLRS201502B of State Key Laboratory of Robotics and System
(HIT).
method, the sufficient conditions for finite-time stability are
presented. We deal with the finite-time stabilization prob-
lem via designing parameter-dependent state feedback con-
trollers. Our resulted conditions are in terms of LMIs. The
effectiveness of the approach is finally illustrated through an
example.
2 Problem Formulation and Preliminaries
The framework of linear parameter varying systems con-
cerns linear dynamical systems where the time dependence
enters the state equations through exogenous parameters. A
state-space description of an LPV system can be represented
as
˙x(t)=A(θ(t))x(t)+B(θ(t))u(t) (1a)
x(t
0
)=x
0
(1b)
where x(t) ∈ R
n
is the state vector, u(t) ∈ R
m
is the control input. The function θ(t)=
θ
1
(t) θ
2
(t) ··· θ
l
(t)
T
∈ Θ is a time-varying
parameter vector, where Θ is the set of allowable parameter
trajectories and is a compact subset of R
l
, i.e., we consider
bounded parameter trajectories. To facilitate the subsequent
developments, the following assumptions are made.
Assumption 1 At each time instant t, the parameter vector
θ(t) is accessible to be measured.
Assumption 2 The state-space matrices A(·) ∈ R
n×n
and
B(·) ∈ R
n×m
are known continuous functions of the param-
eter vector θ(·) ∈ Θ.
Note that θ(t) is measurable from Assumption 1, then we
seek to design controllers that are scheduled based on the
real-time measurement of θ.
In the following, let us extend the definition of finite-time
stability proposed in [21] to LPV systems.
Definition 1 (Finite-Time Stability) Given an initial time t
0
,
positive scalars c
1
, c
2
, T, with c
2
>c
1
, a symmetric positive
definite matrix R, and a given parameter vector θ ∈ Θ, the
LPV system (1) with u =0is said to be finite-time stable
with respect to (t
0
,c
1
,c
2
,T,R,θ) if
x
T
0
Rx
0
≤ c
1
⇒ x
T
(t)Rx(t) <c
2
, (2)
∀t ∈ [t
0
,t
0
+ T ] .
Proceedings of the 35th Chinese Control Conference
Jul
27-29, 2016, Chen
du, China
1646