Anda di halaman 1dari 80

Linear Matrix Inequalities in Control

Guido Herrmann
University of Leicester
Linear Matrix Inequalities in Control p. 1/43
Presentation prepared for

Summerschool at the University
of Leicester, September 2006

Affiliation now:

Department of Mechanical
Engineering, University of Bristol
Presentation
1. Introduction and some simple examples
2. Fundamental Properties and Basic Structure of Linear Matrix Inequalities (LMIs)
3. LMI-problems
4. Tricks in Matrix Inequalities - Approaches to create LMIs from Matrix Inequalities
(a) Congruence Transformation
(b) Change of Variables
(c) Projection Lemma
(d) S-procedure
(e) Schur Complement
5. Examples (L
2
-gain computation, non-linearities, etc)
6. Conclusions
Linear Matrix Inequalities in Control p. 2/43
Introduction - A Simple Example
A linear system
x = Ax
is stable if and only if there is a positive denite P for
V(x) = x
T
Px (i.e. V(x) > 0 f or x = 0)
and
x
T
PAx +x
T
A
T
Px < 0 x = 0
Linear Matrix Inequalities in Control p. 3/43
Introduction - A Simple Example
A linear system
x = Ax
is stable if and only if there is a positive denite P for
V(x) = x
T
Px (i.e. V(x) > 0 f or x = 0)
and
x
T
PAx +x
T
A
T
Px < 0 x = 0
The two matrix inequalities involved here are
PA+A
T
P < 0
and
P > 0.
Linear Matrix Inequalities in Control p. 3/43
Introduction - A Simple Example
A linear system
x = Ax
is stable if and only if there is a positive denite P for
V(x) = x
T
Px (i.e. V(x) > 0 f or x = 0)
and
x
T
PAx +x
T
A
T
Px < 0 x = 0
The two matrix inequalities involved here are
PA+A
T
P < 0
and
P > 0.
The matrix problem here is to nd P so that these inequalities are satised. The
inequalities are linear in P.
Linear Matrix Inequalities in Control p. 3/43
Introduction - LQR-optimal control
We would like to compute a state feedback controller u = Kx controlling
x = Ax +Bu
with an initial condition of x(0) = x
0
.
The cost function
J =
_

0
(x
T
Qx +u
T
Ru)dt
is to be minimized. We know that the solution to this problem is
K = R
1
B
T
P, A
T
P+PA+PBR
1
B
T
P+Q = 0
and J = min
u
_

0
(x
T
Qx +u
T
Ru)dt = x
T
0
Px
0
.
How can we express this problem in terms of an LMI?
Linear Matrix Inequalities in Control p. 4/43
Introduction
In control the requirements for controller design are usually
1. Closed Loop Stability
2. Robustness
3. Performance
4. Robust Performance
Control design requirements are usually best encoded in form of an optimization
criterion
(e.g. robustness in terms of L
2
/small gain-requirements, performance via linear
quadratic control, H

-requirements etc.)
Linear Matrix Inequalities in Control p. 5/43
Introduction
We have seen that stability of a linear autonomous system can be easily expressed via
a linear matrix inequality
We will see that linear quadratic control problems can be expressed in terms of LMIs
L
2
or H

analysis/design problems can be expressed as LMI-problems


Some classes of nonlinearities are easily captured via matrix inequalities
This creates a synergy which allows to express a control design problem via different
seemingly contradictive requirements
For LMIs, very reliable numerical solution tools are available
Linear Matrix Inequalities in Control p. 6/43
Fundamental LMI properties
A matrix Q is dened to be positive denite if it is symmetric and
x
T
Qx > 0 x = 0
This is signied by
Q > 0
Linear Matrix Inequalities in Control p. 7/43
Fundamental LMI properties
A matrix Q is dened to be positive denite if it is symmetric and
x
T
Qx > 0 x = 0
This is signied by
Q > 0
Likewise, Q is said to be positive semi-denite if it is symmetric and
x
T
Qx 0 x thus Q 0
Linear Matrix Inequalities in Control p. 7/43
Fundamental LMI properties
A matrix Q is dened to be positive denite if it is symmetric and
x
T
Qx > 0 x = 0
This is signied by
Q > 0
Likewise, Q is said to be positive semi-denite if it is symmetric and
x
T
Qx 0 x thus Q 0
A matrix Q is negative denite if it is symmetric and
x
T
Qx < 0 x = 0 thus Q < 0
or negative semi-denite if it is symmetric and
x
T
Qx 0 x = 0 thus Q 0
Linear Matrix Inequalities in Control p. 7/43
The Basic Structure of an LMI
Any linear matrix inequality (LMI) can be easily rewritten as
F(v) = F
0
+
m

i=1
v
i
F
i
> 0
where v R
m
is a variable and F
0
, F
i
are given constant symmetric matrices.
This matrix inequality is linear in the variables v
i
.
Linear Matrix Inequalities in Control p. 8/43
The Basic Structure of an LMI
Any linear matrix inequality (LMI) can be easily rewritten as
F(v) = F
0
+
m

i=1
v
i
F
i
> 0
where v R
m
is a variable and F
0
, F
i
are given constant symmetric matrices.
This matrix inequality is linear in the variables v
i
.
For instance for the simple linear matrix inequality in the symmetric P
PA+A
T
P < 0
the variables v R
m
are dened via P R
nn
. Hence, m =
n(n+1)
2
in this case!
Linear Matrix Inequalities in Control p. 8/43
The Basic Structure of an LMI
Another very generic way of writing down an LMI is
F(V
1
,V
2
, . . . ,V
n
) = F
0
+G
1
V
1
H
1
+G
2
V
2
H
2
+. . .
= F
0
+
n

i=1
G
i
V
i
H
i
> 0
where the unstructured V
i
R
q
i
p
i
are matrix variables,

n
i=1
q
i
p
i
= m. We seek to nd
V
i
as they are variables.
The matrices F
0
, G
i
, H
i
are given.
From now on, we will mainly consider LMIs of this form.
Linear Matrix Inequalities in Control p. 9/43
System of LMIs
A system of LMIs is
F
1
(V
1
, . . . ,V
n
) > 0
.
.
. > 0
F
p
(V
1
, . . . ,V
n
) > 0
where
F
j
(V
1
, . . . ,V
n
) = F
0j
+
n

i=1
G
i j
V
i
H
i j
This can be easily changed into a single LMI ...
Linear Matrix Inequalities in Control p. 10/43
System of LMIs
Lets dene

F
0
,

G
i
,

H
i
,

V
i
as

F
0
=
_

_
F
01
0 0 0
0 F
02
0 0
0 0 . . . 0
0 0 0 F
0p
_

_
= diag(F
01
, . . . , F
0p
)

G
i
= diag(G
i1
, . . . , G
ip
)

H
i
= diag(H
i1
, . . . , H
ip
)

V
i
= diag(V
i
, . . .V
i
)
Linear Matrix Inequalities in Control p. 11/43
System of LMIs
Lets dene

F
0
,

G
i
,

H
i
,

V
i
as

F
0
=
_

_
F
01
0 0 0
0 F
02
0 0
0 0 . . . 0
0 0 0 F
0p
_

_
= diag(F
01
, . . . , F
0p
)

G
i
= diag(G
i1
, . . . , G
ip
)

H
i
= diag(H
i1
, . . . , H
ip
)

V
i
= diag(V
i
, . . .V
i
)
We then have the inequality
F
big
(V
1
, . . . ,V
n
) :=

F
0
+
n

i=1

G
i

V
i

H
i
> 0
which is just one single LMI.
But be aware that this time the new variable

V
i
is structured, i.e. not all elements of

V
i
are
free parameters!
Linear Matrix Inequalities in Control p. 11/43
Different classes of LMI-problems: Feasibility Problem
We seek a feasible solution {V
1
, . . . ,V
n
} such that
F(V
1
, . . . ,V
n
) > 0
We are not interested in the optimality of the solution, only in nding a solution, which
satises the LMI and may not be unique.
Linear Matrix Inequalities in Control p. 12/43
Different classes of LMI-problems: Feasibility Problem
We seek a feasible solution {V
1
, . . . ,V
n
} such that
F(V
1
, . . . ,V
n
) > 0
We are not interested in the optimality of the solution, only in nding a solution, which
satises the LMI and may not be unique.
Example: A linear system
x = Ax
is stable if and only if there is a matrix P satisfying
PA+A
T
P < 0
and
P > 0.
Linear Matrix Inequalities in Control p. 12/43
LMI-problems: Linear Objective Minimization
Minimization (or maximization) of a linear scalar function, (.), of the matrix variables V
i
,
subject to LMI constraints:
min(V
1
, . . . ,V
n
) s.t.
..
such that

, subject to

F(V
1
, . . . ,V
n
) > 0
where we have used the abbreviation s.t. to mean such that.
Linear Matrix Inequalities in Control p. 13/43
LMI-problems: Linear Objective Minimization
Minimization (or maximization) of a linear scalar function, (.), of the matrix variables V
i
,
subject to LMI constraints:
min(V
1
, . . . ,V
n
) s.t.
..
such that

, subject to

F(V
1
, . . . ,V
n
) > 0
where we have used the abbreviation s.t. to mean such that.
Example: Calculating the H

norm of a linear system.


x = Ax +Bw
z = Cx +Dw
the H

norm of the transfer function matrix T


zw
from w to z is computed by:
min s.t.
_

_
A
T
P+PA PB C
T
B
T
P I D
T
C D I
_

_
< 0, P > 0.
The LMI variables are P and ! The value of is unique, P is not.
Linear Matrix Inequalities in Control p. 13/43
LMI-problems: Generalized eigenvalue problem
min s.t. F
1
(V
1
, . . . ,V
n
) +F
2
(V
1
, . . . ,V
n
) < 0
F
2
(V
1
, . . . ,V
n
) < 0
F
3
(V
1
, . . . ,V
n
) < 0
Note that in some cases, a GEVP problem can be reduced to a linear objective
minimization problem, through an appropriate change of variables.
Linear Matrix Inequalities in Control p. 14/43
LMI-problems: Generalized eigenvalue problem
min s.t. F
1
(V
1
, . . . ,V
n
) +F
2
(V
1
, . . . ,V
n
) < 0
F
2
(V
1
, . . . ,V
n
) < 0
F
3
(V
1
, . . . ,V
n
) < 0
Note that in some cases, a GEVP problem can be reduced to a linear objective
minimization problem, through an appropriate change of variables.
Example: Bounding the decay rate of a linear system.
The decay rate is the largest such that
x(t) exp(t)x(0), >, x(t)
Lets choose the Lyapunov function V(x) = x
T
Px > 0 and ensure that

V(x) 2V(x).
The problem of nding the decay rate could be posed as
min s.t. A
T
P+PA+2P < 0,
P < 0,
Linear Matrix Inequalities in Control p. 14/43
LMI-problems: Generalized eigenvalue problem
min s.t. F
1
(V
1
, . . . ,V
n
) +F
2
(V
1
, . . . ,V
n
) < 0
F
2
(V
1
, . . . ,V
n
) < 0
F
3
(V
1
, . . . ,V
n
) < 0
Note that in some cases, a GEVP problem can be reduced to a linear objective
minimization problem, through an appropriate change of variables.
Example: Bounding the decay rate of a linear system.
The decay rate is the largest such that
x(t) exp(t)x(0), >, x(t)
Lets choose the Lyapunov function V(x) = x
T
Px > 0 and ensure that

V(x) 2V(x).
The problem of nding the decay rate could be posed as
min s.t. A
T
P+PA+2P < 0, i.e. F
1
(P) := A
T
P+PA
P < 0, i.e. F
2
(P) := 2P
i.e. F
3
(P) := I
Linear Matrix Inequalities in Control p. 14/43
Tricks: Congruence transformation
We know that for Q R
nn
Q > 0
and a real W R
nn
such that rank(W) = n, the following inequality holds
WQW
T
> 0
Deniteness of a matrix is invariant under pre and post-multiplication by a full rank real
matrix, and its transpose, respectively.
Often W is chosen to have a diagonal structure.
Linear Matrix Inequalities in Control p. 15/43
Tricks: Change of variables
By dening new variables, it is sometimes possible to linearise nonlinear MIs
Example: State feedback control synthesis
Find F such that the eigenvalues of A+BF are in the open left-half complex plane
This is equivalent to nding a matrix F and P > for
(A+BF)
T
P+P(A+BF) < 0 or A
T
P+PA+F
T
B
T
P+PBF < 0
! Terms with products of F and P are nonlinear or bilinear !
Linear Matrix Inequalities in Control p. 16/43
Tricks: Change of variables
By dening new variables, it is sometimes possible to linearise nonlinear MIs
Example: State feedback control synthesis
Find F such that the eigenvalues of A+BF are in the open left-half complex plane
This is equivalent to nding a matrix F and P > for
(A+BF)
T
P+P(A+BF) < 0 or A
T
P+PA+F
T
B
T
P+PBF < 0
! Terms with products of F and P are nonlinear or bilinear !
Multiply with Q := P
1
> 0 (A very simple case of congruence transformation):
QA
T
+AQ+QF
T
B
T
+BFQ < 0
This is a new matrix inequality in the variables Q > 0 and F (still non-linear).
Linear Matrix Inequalities in Control p. 16/43
Tricks: Change of variables
QA
T
+AQ+QF
T
B
T
+BFQ < 0
Dene a second new variable L = FQ
QA
T
+AQ+L
T
B
T
+BL < 0
We now have an LMI feasibility problem in the new variables Q > 0 and L.
Recovery of F and P by
F = LQ
1
, P = Q
1
.
Linear Matrix Inequalities in Control p. 17/43
Tricks: Schur complement
Schurs formula says that the following statements are equivalent:
i. =
_

11

12

T
12

22
_
< 0
ii.
22
< 0

11

12

1
22

T
12
< 0
The main use is to transform quadratic matrix inequalities into linear matrix inequalities.
Linear Matrix Inequalities in Control p. 18/43
Tricks: Schur complement
Schurs formula says that the following statements are equivalent:
i. =
_

11

12

T
12

22
_
< 0
ii.
22
< 0

11

12

1
22

T
12
< 0
The main use is to transform quadratic matrix inequalities into linear matrix inequalities.
Example: Making a LQR-type quadratic inequality linear (Riccati inequality)
A
T
P+PA+PBR
1
B
T
P+Q < 0
where P > 0 is the matrix variable and Q, R > 0 are constant.
Linear Matrix Inequalities in Control p. 18/43
Tricks: Schur complement
Schurs formula says that the following statements are equivalent:
i. =
_

11

12

T
12

22
_
< 0
ii.
22
< 0

11

12

1
22

T
12
< 0
The main use is to transform quadratic matrix inequalities into linear matrix inequalities.
Example: Making a LQR-type quadratic inequality linear (Riccati inequality)
A
T
P+PA+PBR
1
B
T
P+Q < 0
where P > 0 is the matrix variable and Q, R > 0 are constant.
The Riccati inequality can be transformed into
_
A
T
P+PA+Q PB
R
_
< 0
Linear Matrix Inequalities in Control p. 18/43
Tricks: Schur complement
Example: Making a LQR-type quadratic inequality linear (Riccati inequality)
A
T
P+PA+PBR
1
B
T
P+Q < 0 is equivalent to
_
A
T
P+PA+Q PB
R
_
< 0
where P > 0 is the matrix variable and Q, R > 0 are constant. This inequality can be used
to minimize the cost function
J =
_

0
(x
T
Qx +u
T
Ru)dt
for the computation of the state feedback controller u = Kx controlling
x = Ax +Bu
with an initial condition of x(0) = x
0
.
Linear Matrix Inequalities in Control p. 19/43
Tricks: Schur complement
Example: Making a LQR-type quadratic inequality linear (Riccati inequality)
A
T
P+PA+PBR
1
B
T
P+Q < 0 is equivalent to
_
A
T
P+PA+Q PB
R
_
< 0
where P > 0 is the matrix variable and Q, R > 0 are constant. This inequality can be used
to minimize the cost function
J =
_

0
(x
T
Qx +u
T
Ru)dt
for the computation of the state feedback controller u = Kx controlling
x = Ax +Bu
with an initial condition of x(0) = x
0
. We know that the solution to this problem is
K = R
1
B
T

P, A
T

P+

PA+

PBR
1
B
T

P+Q = 0
and J = min
u
_

0
(x
T
Qx +u
T
Ru)dt = x
T
0

Px
0
.
Linear Matrix Inequalities in Control p. 19/43
Tricks: Schur complement
The alternative solution to the optimization problem is given by the following
LMI-problem:
minx
T
0
Px
0
s.t.
_
A
T
P+PA+Q PB
R
_
< 0
P < 0
for which the optimal controller is given by K =R
1
B
T
P.
Linear Matrix Inequalities in Control p. 20/43
Tricks: The S-procedure
We would like to guarantee that a single quadratic function of x R
m
is such that
F
0
(x) 0 F
0
(x) := x
T
A
0
x +2b
0
x +c
0
whenever certain other quadratic functions are positive semi-denite
F
i
(x) 0 F
i
(x) := x
T
A
i
x +2b
0
x +c
0
, i {1, 2, . . . , q}
Linear Matrix Inequalities in Control p. 21/43
Tricks: The S-procedure
We would like to guarantee that a single quadratic function of x R
m
is such that
F
0
(x) 0 F
0
(x) := x
T
A
0
x +2b
0
x +c
0
whenever certain other quadratic functions are positive semi-denite
F
i
(x) 0 F
i
(x) := x
T
A
i
x +2b
0
x +c
0
, i {1, 2, . . . , q}
Illustration:
Consider i = 1. We need to ensure F
0
(x) 0 for all x such that F
1
(x) 0.
If there is a scalar, > 0, such that
F
aug
(x) := F
0
(x) +F
1
(x) 0 x s.t.F
1
(x) 0
then our goal is achieved.
F
aug
(x) 0 implies that F
0
(x) 0 if F
1
(x) 0 because F
0
(x) F
aug
(x) if F
1
(x) 0.
Linear Matrix Inequalities in Control p. 21/43
Tricks: The S-procedure
F
aug
(x) := F
0
(x) +F
1
(x) 0 x s.t.F
1
(x) 0
Extending this idea to q inequality constraints:
F
0
(x) 0 whenever F
i
(x) 0 ()
holds if
F
0
(x) +
q

i=1

i
F
i
(x) 0,
i
0 ()
Linear Matrix Inequalities in Control p. 22/43
Tricks: The S-procedure
F
aug
(x) := F
0
(x) +F
1
(x) 0 x s.t.F
1
(x) 0
Extending this idea to q inequality constraints:
F
0
(x) 0 whenever F
i
(x) 0 ()
holds if
F
0
(x) +
q

i=1

i
F
i
(x) 0,
i
0 ()
The S-procedure is conservative; inequality () implies inequality ()
Equivalence is only guaranteed for i = 1.
The
i
s are usually variables in an LMI problem.
Linear Matrix Inequalities in Control p. 22/43
Tricks: The Projection Lemma
We sometimes encounter inequalities of the form
(V) +G(V)H
T
(V) +H(V)
T
G
T
(V) < 0 ()
where V and are the matrix variables, is an unstructured matrix variable.
(.), G(.), H(.) are (normally afne) functions of V.
Linear Matrix Inequalities in Control p. 23/43
Tricks: The Projection Lemma
We sometimes encounter inequalities of the form
(V) +G(V)H
T
(V) +H(V)
T
G
T
(V) < 0 ()
where V and are the matrix variables, is an unstructured matrix variable.
(.), G(.), H(.) are (normally afne) functions of V.
Inequality () is satised for some V if and only if
_
W
T
G(V)
(V)W
G(V)
< 0
W
T
H(V)
(V)W
H(V)
< 0
Linear Matrix Inequalities in Control p. 23/43
Tricks: The Projection Lemma
We sometimes encounter inequalities of the form
(V) +G(V)H
T
(V) +H(V)
T
G
T
(V) < 0 ()
where V and are the matrix variables, is an unstructured matrix variable.
(.), G(.), H(.) are (normally afne) functions of V.
Inequality () is satised for some V if and only if
_
W
T
G(V)
(V)W
G(V)
< 0
W
T
H(V)
(V)W
H(V)
< 0
where W
G(V)
and W
H(V)
are the orthogonal complements of G(V) and H(V), i.e.
W
G(V)
G(V) = 0 W
H(V)
H(V) = 0.
and [W
T
G(V)
G(V)], [W
T
H(V)
H(V)] are both full rank.
Linear Matrix Inequalities in Control p. 23/43
Tricks: The Projection Lemma
The main point is that we can transform a matrix inequality which is a function of two
variables, V and , into two inequalities which are functions of one variable:
(i) It can facilitate the derivation of an LMI.
(ii) There are less variables for computation.
Linear Matrix Inequalities in Control p. 24/43
Tricks: The Projection Lemma
The main point is that we can transform a matrix inequality which is a function of two
variables, V and , into two inequalities which are functions of one variable:
(i) It can facilitate the derivation of an LMI.
(ii) There are less variables for computation.
It is often the approach is to solve for V using
_
W
T
G(V)
(V)W
G(V)
< 0
W
T
H(V)
(V)W
H(V)
< 0
and then for using
(V) +G(V)H
T
(V) +H(V)
T
G
T
(V) < 0
Note that this can be numerically unreliable!!
Linear Matrix Inequalities in Control p. 24/43
Examples: L
2
gain- Continuous-time systems
Linear systems: The H

norm is equivalent to the maximum RMS (Root-Mean-Square)


energy gain, the H

-gain of a linear system, the L


2
gain.
Linear Matrix Inequalities in Control p. 25/43
Examples: L
2
gain- Continuous-time systems
Linear systems: The H

norm is equivalent to the maximum RMS (Root-Mean-Square)


energy gain, the H

-gain of a linear system, the L


2
gain.
A system with input w(t) and output z(t) is said to have an L
2
gain of if
z
2
<w
2
+, > 0
where w
2
=
_
_

t=0
w

(t)w(t)dt.
The L
2
gain is a measure of the output relative to the size of its input.
Linear Matrix Inequalities in Control p. 25/43
Examples: L
2
gain- Continuous-time systems
Linear systems: The H

norm is equivalent to the maximum RMS (Root-Mean-Square)


energy gain, the H

-gain of a linear system, the L


2
gain.
A system with input w(t) and output z(t) is said to have an L
2
gain of if
z
2
<w
2
+, > 0
where w
2
=
_
_

t=0
w

(t)w(t)dt.
The L
2
gain is a measure of the output relative to the size of its input.
The H

norm of x = Ax +Bw, z =Cx +Dw is given by:


min s.t.
_

_
A
T
P+PA PB C
T
B
T
P I D
T
C D I
_

_
< 0, P > 0.
Linear Matrix Inequalities in Control p. 25/43
Examples: L
2
gain- Continuous-time systems
_

_
A
T
P+PA PB C
T
B
T
P I D
T
C D I
_

_
< 0, P > 0.
The Schur complement gives
_
A
T
P+PA+
1

C
T
C PB+
1

C
T
D
B
T
P+
1

D
T
C I +
1

D
T
D
_
Linear Matrix Inequalities in Control p. 26/43
Examples: L
2
gain- Continuous-time systems
_

_
A
T
P+PA PB C
T
B
T
P I D
T
C D I
_

_
< 0, P > 0.
The Schur complement gives
_
A
T
P+PA+
1

C
T
C PB+
1

C
T
D
B
T
P+
1

D
T
C I +
1

D
T
D
_
In terms of
_
x
T
w
T

T
, it follows that we need to nd the minimum of so that
_
x
w
_
T
_
A
T
P+PA+
1

C
T
C PB+
1

C
T
D
B
T
P+
1

D
T
C I +
1

D
T
D
__
x
w
_
< 0,
_
x
T
w
T

T
= 0
Linear Matrix Inequalities in Control p. 26/43
Examples: L
2
gain- Continuous-time systems
_

_
A
T
P+PA PB C
T
B
T
P I D
T
C D I
_

_
< 0, P > 0.
The Schur complement gives
_
A
T
P+PA+
1

C
T
C PB+
1

C
T
D
B
T
P+
1

D
T
C I +
1

D
T
D
_
In terms of
_
x
T
w
T

T
, it follows that we need to nd the minimum of so that
_
x
w
_
T
_
A
T
P+PA+
1

C
T
C PB+
1

C
T
D
B
T
P+
1

D
T
C I +
1

D
T
D
__
x
w
_
< 0,
_
x
T
w
T

T
= 0
or
x
T
A
T
Px +x
T
PAx +
1

x
T
C
T
Cx +x
T
(PB+
1

C
T
D)w+w
T
(B
T
P+
1

D
T
C)x +w
T
1

D
T
Dww
T
w
= x
T
A
T
Px +x
T
PA
T
x +2x
T
PBw+
1

z
T
z w
T
w < 0
Linear Matrix Inequalities in Control p. 26/43
Examples: L
2
gain- Continuous-time systems
Dening V = x
T
Px

V = x
T
A
T
Px +x
T
PAx +2x
T
PBw
Thus, we require:
x
T
A
T
Px +x
T
PA
T
x +2x
T
PBw+
1

z
T
z w
T
w =

V +
1

z
T
z w
T
w < 0
Linear Matrix Inequalities in Control p. 27/43
Examples: L
2
gain- Continuous-time systems
Dening V = x
T
Px

V = x
T
A
T
Px +x
T
PAx +2x
T
PBw
Thus, we require:
x
T
A
T
Px +x
T
PA
T
x +2x
T
PBw+
1

z
T
z w
T
w =

V +
1

z
T
z w
T
w < 0
and integration in the interval [0, ) implies
V(t =) V(t = 0) +
_

t=0
1

z
T
(s)z(s)ds
_

t=0
w
T
(s)w(s)ds < 0
_

t=0
z
T
(s)z(s)ds <
_

t=0

2
w
T
(s)w(s)ds +(V(t = 0) V(t =))
_
_

t=0
z
T
(s)z(s)ds <
_
_

t=0

2
w
T
(s)w(s)ds +V(t = 0)
_
_

t=0
z
T
(s)z(s)ds <
_
_

t=0
w
T
(s)w(s)ds +
_
V(t = 0)
Linear Matrix Inequalities in Control p. 27/43
Examples: L
2
gain- Continuous-time systems
V(t =) V(t = 0) +
_

t=0
1

z
T
(s)z(s)ds
_

t=0
w
T
(s)w(s)ds < 0
_
_

t=0
z
T
(s)z(s)ds <
_
_

t=0
w
T
(s)w(s)ds+
_
V(t = 0)
z
2
< w
2
+
..

V(t=0)
Thus, the linear system (A, B,C, D) has indeed an L
2
gain .
Linear Matrix Inequalities in Control p. 28/43
Examples: Discrete-time systems
A linear discrete system
x(k +1) = Ax(k)
is asymptotically stable if and only if there is
V(x) = x
T
Px, P > 0.
and
V(x(k +1)) V(x(k)) =V(x(k +1)) = x
T
(k)A
T
PAx(k) x
T
(k)Px(k) < 0 x(k) = 0
or
A
T
PAP < 0.
Linear Matrix Inequalities in Control p. 29/43
Examples: l
2
gain- Discrete-time systems
A system with input w(t) and output z(t) is said to have an L
2
gain of if
z
2
<w
2
+, > 0
where w
2
=
_

k=0
w
T
(k)w(k) .
Linear Matrix Inequalities in Control p. 30/43
Examples: l
2
gain- Discrete-time systems
A system with input w(t) and output z(t) is said to have an L
2
gain of if
z
2
<w
2
+, > 0
where w
2
=
_

k=0
w
T
(k)w(k) .
For linear systems
x(k +1) = Ax(k) +Bw(k)
y =Cx(k) +Dw(k)
the value of the nite l
2
-gain, , (H

norm; the maximum RMS energy gain) is:


min s.t.
_
A
T
PAP+
1

C
T
C A
T
PB+
1

C
T
D
1

D
T
C I +B
T
PB+
1

D
T
D
_
< 0
P < 0
for P > 0.
Linear Matrix Inequalities in Control p. 30/43
Examples: l
2
gain- Discrete-time systems
The l
2
gain relationship readily follows for V = x
T
Px from:
V(x(k +1)) +
1

y
T
(k)y(k) w
T
(k)w(k) < 0

k=0
V(x(k +1))
. .
V(x())V(x(0))
+
1

k=0
y
T
(k)y(k)

k=0
w
T
(k)w(k) < 0
Linear Matrix Inequalities in Control p. 31/43
Examples: l
2
gain- Discrete-time systems
The l
2
gain relationship readily follows for V = x
T
Px from:
V(x(k +1)) +
1

y
T
(k)y(k) w
T
(k)w(k) < 0

k=0
V(x(k +1))
. .
V(x())V(x(0))
+
1

k=0
y
T
(k)y(k)

k=0
w
T
(k)w(k) < 0
Problem : The matrix inequality
_
A
T
PAP+
1

C
T
C A
T
PB+
1

C
T
D
1

D
T
C I +B
T
PB+
1

D
T
D
_
< 0
is not linear. P > 0 and > 0 are variables.
Linear Matrix Inequalities in Control p. 31/43
Examples: l
2
gain- Discrete-time systems
The l
2
gain relationship readily follows for V = x
T
Px from:
V(x(k +1)) +
1

y
T
(k)y(k) w
T
(k)w(k) < 0

k=0
V(x(k +1))
. .
V(x())V(x(0))
+
1

k=0
y
T
(k)y(k)

k=0
w
T
(k)w(k) < 0
Problem : The matrix inequality
_
A
T
PAP+
1

C
T
C A
T
PB+
1

C
T
D
1

D
T
C I +B
T
PB+
1

D
T
D
_
< 0
is not linear. P > 0 and > 0 are variables.
The Schur Complement implies
_

_
A
T
PAP A
T
PB C
T
B
T
PA I +B
T
PB D
T
C D I
_

_
< 0
Linear Matrix Inequalities in Control p. 31/43
Examples: l
2
gain- Discrete-time systems
_
A
T
PAP+
1

C
T
C A
T
PB+
1

C
T
D
1

D
T
C I +B
T
PB+
1

D
T
D
_
< 0
Congruence transformation & Change of variable approach:
_
A
T
PAP+C
T
C A
T
PB+C
T
D
D
T
C
2
I +B
T
PB+D
T
D
_
< 0
Dening Q = P and =
2
:
min s.t.
_
A
T
QAQ+C
T
C A
T
QB+C
T
D
D
T
C I +B
T
QB+D
T
D
_
< 0
Q < 0
Q > 0 and the scalar > 0 are variables.
The l
2
-gain is readily computed with =

.
Linear Matrix Inequalities in Control p. 32/43
Examples: Sector boundedness
The saturation function is dened as
sat(u) = [sat
1
(u
1
), . . . , sat
2
(u
m
)]
T
and sat
i
(u
i
) = sign(u
i
) min{|u
i
|, u
i
}, u
i
> 0 i {1, . . . , m}
u
i
is the ith saturation limit
Linear Matrix Inequalities in Control p. 33/43
Examples: Sector boundedness
The saturation function is dened as
sat(u) = [sat
1
(u
1
), . . . , sat
2
(u
m
)]
T
and sat
i
(u
i
) = sign(u
i
) min{|u
i
|, u
i
}, u
i
> 0 i {1, . . . , m}
u
i
is the ith saturation limit
It is easy to verify that the saturation function, sat
i
(u
i
) satises the following inequality
u
i
sat
i
(u
i
) sat
2
i
(u
i
)
Linear Matrix Inequalities in Control p. 33/43
Examples: Sector boundedness
It is easy to verify that the saturation function, sat
i
(u
i
) satises the following inequality
u
i
sat
i
(u
i
) sat
2
i
(u
i
)
or
sat
i
(u
i
)[u
i
sat
i
(u
i
)]w
i
0
for some w
i
> 0.
Linear Matrix Inequalities in Control p. 34/43
Examples: Sector boundedness
It is easy to verify that the saturation function, sat
i
(u
i
) satises the following inequality
u
i
sat
i
(u
i
) sat
2
i
(u
i
)
or
sat
i
(u
i
)[u
i
sat
i
(u
i
)]w
i
0
for some w
i
> 0. We can write
sat(u)
T
W[usat(u)] 0
for some diagonal W = diag(w
1
, w
2
, . . .) > 0.
This inequality can be easily used in an S procedure approach where the elements of
the diagonal matrix W act as free parameters if necessary/possible.
Linear Matrix Inequalities in Control p. 34/43
Examples: A slightly more detailed example
Consider the L
2
-gain for the SISO-system with saturated input signal u:
x = Ax +bsat(u), x R
n
and a limited measurement range of the output y:
y = sat(cx +dsat(u)).
Linear Matrix Inequalities in Control p. 35/43
Examples: A slightly more detailed example
Consider the L
2
-gain for the SISO-system with saturated input signal u:
x = Ax +bsat(u), x R
n
and a limited measurement range of the output y:
y = sat(cx +dsat(u)).
The limits at the actuator inputs u can be due to mechanical limits (e.g. valves) or due to
digital-to-analogue converter voltage signal limits.
Output signals can be constrained due to sensor voltage range limits or simply by
analogue-to-digital converter limits.
The analysis of such systems is vital to practical control systems and will be pursued in
greater detail later. We may consider here an L
2
gain analysis.
Linear Matrix Inequalities in Control p. 35/43
Examples: A slightly more detailed example
We may dene
s = sat(u).
Hence, it follows
sw
1
(us) 0, w
1
> 0
For the output signal y:
yw
2
(cx +ds y) 0, w
2
> 0
Linear Matrix Inequalities in Control p. 36/43
Examples: A slightly more detailed example
We may dene
s = sat(u).
Hence, it follows
sw
1
(us) 0, w
1
> 0
For the output signal y:
yw
2
(cx +ds y) 0, w
2
> 0
We know that from

V +
1

y
2
u
2
0
follows that our system has the L
2
-gain .
We have to consider the two saturation nonlinearities!
Linear Matrix Inequalities in Control p. 36/43
Examples: A slightly more detailed example
With the S-procedure

V +
1

y
2
u
2
+2sw
1
(us) +2yw
2
(cx +ds y) < 0 f or
_
x
T
u y
_
= 0
the system has also an L
2
-gain of .
The expression

V implies:
x
T
A
T
Px +xPAx +x
T
Pbs +sb
T
Px
+
1

y
2
u
2
+2sw
1
(us) +2yw
2
(cx +ds y) 0
Linear Matrix Inequalities in Control p. 37/43
Examples: A slightly more detailed example
With the S-procedure

V +
1

y
2
u
2
+2sw
1
(us) +2yw
2
(cx +ds y) < 0 f or
_
x
T
u y
_
= 0
the system has also an L
2
-gain of .
The expression

V implies:
x
T
A
T
Px +xPAx +x
T
Pbs +sb
T
Px
+
1

y
2
u
2
+2sw
1
(us) +2yw
2
(cx +ds y) 0
Rewriting:
_

_
x
s
u
y
_

_
T
_

_
A
T
P+PA Pb 0 c
T
w
2
b
T
P 2w
1
w
1
dw
2
0 w
1
0
w
2
c w
2
d 0 2w
2
+
1

_
_

_
x
s
u
y
_

_
< 0
for
_
x
T
s u y
_
= 0.
Linear Matrix Inequalities in Control p. 37/43
Examples: A slightly more detailed example
This is equivalent to
_

_
A
T
P+PA Pb 0 c
T
w
2
b
T
P 2w
1
w
1
dw
2
0 w
1
0
w
2
c w
2
d 0 2w
2
+
1

_
< 0.
We would like to minimize , while P, w
1
, w
2
are variables. Not an LMI!
Linear Matrix Inequalities in Control p. 38/43
Examples: A slightly more detailed example
This is equivalent to
_

_
A
T
P+PA Pb 0 c
T
w
2
b
T
P 2w
1
w
1
dw
2
0 w
1
0
w
2
c w
2
d 0 2w
2
+
1

_
< 0.
We would like to minimize , while P, w
1
, w
2
are variables. Not an LMI!
Using Projection Lemma twice, we can derive a signicantly simpler matrix inequality
which delivers the L
2
-gain.
First Step:
_

_
A
T
P+PA Pb 0 c
T
w
2
b
T
P 0 0 dw
2
0 0 0
w
2
c w
2
d 0 2w
2
+
1

_
+
_

_
0
1
0
0
_

_
w
1
_
0 1 1 0
_
+
_

_
0
1
1
0
_

_
w
1
_
0 1 0 0
_
< 0.
Linear Matrix Inequalities in Control p. 38/43
Examples: A slightly more detailed example
First Step:
_

_
A
T
P+PA Pb 0 c
T
w
2
b
T
P 0 0 dw
2
0 0 0
w
2
c w
2
d 0 2w
2
+
1

_
+
_

_
0
1
0
0
_

_
w
1
_
0 1 1 0
_
+
_

_
0
1
1
0
_

_
w
1
_
0 1 0 0
_
< 0.
Dening the matrices
g
1
=
_

_
0
1
0
0
_

_
, h
1
=
_

_
0
1
1
0
_

_
,
1
=
_

_
A
T
P+PA Pb 0 c
T
w
2
b
T
P 0 0 dw
2
0 0 0
w
2
c w
2
d 0 2w
2
+
1

_
,
allows us to write
1
+g
1
w
1
h
T
1
+h
1
w
T
1
g
T
1
< 0.
Linear Matrix Inequalities in Control p. 39/43
Examples: A slightly more detailed example
g
1
=
_

_
0
1
0
0
_

_
, h
1
=
_

_
0
1
1
0
_

_
,
1
=
_

_
A
T
P+PA Pb 0 c
T
w
2
b
T
P 0 0 dw
2
0 0 0
w
2
c w
2
d 0 2w
2
+
1

_
,
The null space matrices W
g
1
and W
h
1
satisfy
_
W
T
g
1
g
1

&
_
W
T
h
1
h
1
_
f ull rank; W
g
1
g
1
= 0, W
h
1
h
1
= 0
Linear Matrix Inequalities in Control p. 40/43
Examples: A slightly more detailed example
g
1
=
_

_
0
1
0
0
_

_
, h
1
=
_

_
0
1
1
0
_

_
,
1
=
_

_
A
T
P+PA Pb 0 c
T
w
2
b
T
P 0 0 dw
2
0 0 0
w
2
c w
2
d 0 2w
2
+
1

_
,
The null space matrices W
g
1
and W
h
1
satisfy
_
W
T
g
1
g
1

&
_
W
T
h
1
h
1
_
f ull rank; W
g
1
g
1
= 0, W
h
1
h
1
= 0
Hence,
W
g
1
=
_

_
I 0 0 0
0 0 1 0
0 0 0 1
_

_
, W
h
1
=
_

_
I 0 0 0
0 1 1 0
0 0 0 1
_

_
Linear Matrix Inequalities in Control p. 40/43
Examples: A slightly more detailed example
g
1
=
_

_
0
1
0
0
_

_
, h
1
=
_

_
0
1
1
0
_

_
,
1
=
_

_
A
T
P+PA Pb 0 c
T
w
2
b
T
P 0 0 dw
2
0 0 0
w
2
c w
2
d 0 2w
2
+
1

_
,
The null space matrices W
g
1
and W
h
1
satisfy
_
W
T
g
1
g
1

&
_
W
T
h
1
h
1
_
f ull rank; W
g
1
g
1
= 0, W
h
1
h
1
= 0
Hence,
W
g
1
=
_

_
I 0 0 0
0 0 1 0
0 0 0 1
_

_
, W
h
1
=
_

_
I 0 0 0
0 1 1 0
0 0 0 1
_

_
Hence, it follows
W
g
1

1
W
T
g
1
=
_

_
A
T
P+PA 0 c
T
w
2
0 0
w
2
c 0 2w
2
+
1

_
, W
h
1

1
W
T
h
1
=
_

_
A
T
P+PA Pb c
T
w
2
b
T
P dw
2
w
2
c w
2
d 2w
2
+
1

_
Linear Matrix Inequalities in Control p. 40/43
Examples: A slightly more detailed example
W
g
1

1
W
T
g
1
=
_

_
A
T
P+PA 0 c
T
w
2
0 0
w
2
c 0 2w
2
+
1

_
, W
h
1

1
W
T
h
1
=
_

_
A
T
P+PA Pb c
T
w
2
b
T
P dw
2
w
2
c w
2
d 2w
2
+
1

_
If W
h
1

1
W
T
h
1
< 0 then also W
g
1

1
W
T
g
1
< 0 (easily seen from a further analysis using the
Projection lemma).
We may carry on investigating W
h
1

1
W
T
h
1
only
Linear Matrix Inequalities in Control p. 41/43
Examples: A slightly more detailed example
W
g
1

1
W
T
g
1
=
_

_
A
T
P+PA 0 c
T
w
2
0 0
w
2
c 0 2w
2
+
1

_
, W
h
1

1
W
T
h
1
=
_

_
A
T
P+PA Pb c
T
w
2
b
T
P dw
2
w
2
c w
2
d 2w
2
+
1

_
If W
h
1

1
W
T
h
1
< 0 then also W
g
1

1
W
T
g
1
< 0 (easily seen from a further analysis using the
Projection lemma).
We may carry on investigating W
h
1

1
W
T
h
1
only
W
h
1

1
W
T
h
1
=
2
+g
2
w
2
h
T
2
+h
2
w
2
g
T
2
where
g
2
=
_

_
0
0
1
_

_
, h
2
=
_

_
c
T
d
1
_

_
,
2
=
_

_
A
T
P+PA Pb 0
b
T
P 0
0 0
1

_
This allows us to derive the null space matrices W
g
2
and W
h
2
for g
2
and h
2
W
g
2
=
_
I 0 0
0 1 0
_
, W
h
2
=
_
I 0 c
T
0 1 d
_
Linear Matrix Inequalities in Control p. 41/43
Examples: A slightly more detailed example
W
g
2
=
_
I 0 0
0 1 0
_
, W
h
2
=
_
I 0 c
T
0 1 d
_
,
2
=
_

_
A
T
P+PA Pb 0
b
T
P 0
0 0
1

_
Thus,
W
g
2

2
W
T
g
2
=
_
A
T
P+PA Pb
b
T
P
_
, W
h
2

2
W
T
h
2
=
_
A
T
P+PA+
c
T
c

Pb+
c
T
d

b
T
P+
dc

+
d
2

_
.
Linear Matrix Inequalities in Control p. 42/43
Examples: A slightly more detailed example
W
g
2
=
_
I 0 0
0 1 0
_
, W
h
2
=
_
I 0 c
T
0 1 d
_
,
2
=
_

_
A
T
P+PA Pb 0
b
T
P 0
0 0
1

_
Thus,
W
g
2

2
W
T
g
2
=
_
A
T
P+PA Pb
b
T
P
_
, W
h
2

2
W
T
h
2
=
_
A
T
P+PA+
c
T
c

Pb+
c
T
d

b
T
P+
dc

+
d
2

_
.
W
g
2

2
W
T
g
2
< 0 is always satised if W
h
2

2
W
T
h
2
.
Hence, the L
2
gain is computed using
_
A
T
P+PA+
c
T
c

Pb+
c
T
d

b
T
P+
dc

+
d
2

_
< 0, P > 0
The L
2
-gain of the linear system (A, b, c, d) is an upper bound of the non-linear operator.
The L
2
-gain of the non-linear and linear operator are identical.
Linear Matrix Inequalities in Control p. 42/43
Summary
Matrix inequalities have shown to be versatile tool to
1. represent L
2
, H

, linear quadratic performance constraints, H


2
etc.
2. analyze linear parameter varying systems, mild non-linear systems
3. combine several analysis problems in one frame work
Matrix inequalities can often be transformed into linear matrix inequalities
by congruence transformation, change of variable approach, etc.
Existence of a large variety of powerful tools for solving LMIs (semi-denite
programming)
LMIs have become a standard tool in the analysis and controller design of linear and
non-linear control systems
Linear Matrix Inequalities in Control p. 43/43

Anda mungkin juga menyukai