Control Systems I

Lecture 4: Diagonalization, Modal Analysis, Intro to Feedback

Readings:

Emilio Frazzoli

Institute for Dynamic Systems and Control

D-MAVT
ETH Zürich

October 13, 2017

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 1 / 26

Tentative schedule

# Date Topic
1 Sept. 22 Introduction, Signals and Systems
2 Sept. 29 Modeling, Linearization
3 Oct. 6 Analysis 1: Time response, Stability
4 Oct. 13 Analysis 2: Diagonalization, Modal coordi-
nates.
5 Oct. 20 Transfer functions 1: Definition and properties
6 Oct. 27 Transfer functions 2: Poles and Zeros
7 Nov. 3 Analysis of feedback systems: internal stability,
root locus
8 Nov. 10 Frequency response
9 Nov. 17 Analysis of feedback systems 2: the Nyquist
condition
10 Nov. 24 Specifications for feedback systems
11 Dec. 1 Loop Shaping
12 Dec. 8 PID control
13 Dec. 15 Implementation issues
14 Dec. 22 Robustness

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 2 / 26

Recap of the previous lecture
LTI system:
ẋ(t) = Ax(t) + Bu(t),
y (t) = Cx(t) + Du(t).
Time response: Z t
x(t) = e At x0 + e A(t−τ ) Bu(τ ) dτ,
0
Z t
y (t) = Ce At x0 + C e A(t−τ ) Bu(τ ) dτ + Du(t).
0
Easy to compute if the matrix A is diagonal, in which case:
Z t
xi (t) = e λi t x0,i + e λi (t−τ ) bi u(τ ) dτ,
0
n
X
y (t) = ci xi (t) + Du(t).
i=1

The eigenvalues of A, λi , can be real or complex-conjugate, giving rise to simple

exponentials, or oscillations with exponentially changing magnitude, respectively.
Asymptotically stable if Re(λi ) < 0 for all i.

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 3 / 26

Today’s learning objectives

After today’s lecture, you should be able to:

Diagonalize a matrix using similarity transformations.

Describe the behavior of an LTI system in modal coordinates.

Understand concepts like controllability and observability.

Understand the basic effects of feedback control on the closed-loop dynamics.

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 4 / 26

Similarity Transformations

The choice of a state-space model for a given system is not unique.

For example, let T be an invertible matrix, and consider a coordinate
transpormation x = T x̃, i.e., x̃ = T −1 x. This is called a similarity
transformation.
The standard state-space model can be written as

T x̃˙ = AT x̃ + Bu,
 
ẋ = Ax + Bu,
⇒
y = Cx + Du. y = CT x̃ + Du.

i.e.,

x̃˙ = (T −1 AT )x̃ + (T −1 B)u = Ãx̃ + B̃u

y = (CT )x̃ + Du = C̃ x̃ + D̃u.

You can check that the time response is exactly the same for the two models
(A, B, C , D) and (Ã, B̃, C̃ , D̃)!

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 5 / 26

Diagonalization

Let λi , vi be respectively an eigenvalue and an eigenvector of A, i.e.,

Avi = λi vi .

Now assume we have n (=dim. of x and A) independent eigenvectors; then

we can assemble the eigenvectors into an invertible matrix V whose columns
are the eigenvectors vi . Then
   

AV = A  v1 v2 ... vn  =  λ1 v1 λ 2 v2 ... λn vn  = V Λ.

In other words, if a square matrix A has a full set of independent

eigenvectors, then it is diagonalizable (and vice-versa), with the similarity
transformation given by a matrix whose columns are the eigenvectors.

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 6 / 26

Modal decomposition
The entries in the diagonal matrix à = Λ are the eigenvalues λ1 , . . . , λn of
the matrix A.
Since x̃(t) = e Ãt x̃(0), we get that each component of the homogeneous
solution is given by
x̃i (t) = e λi t x̃i (0).

Furthermore, if x(0) = vi for some i = 1, . . . , n, then by the definition of

matrix exponential and eigenvalues/eigenvectors

x(t) = e At vi = e λi t vi .

If the eigenvectors vi , i = 1, . . . , n form a basis, then we can always express

any initial condition as a linear combination of eigenvectors, i.e.,
x(0) = V x̃(0), with x̃0 = V −1 x0 , and write
n
X
x(t) = e λi t x̃i (0) vi .
i=1

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 7 / 26

Modal coordinates

Eigenvalues and eigenvectors of A define the modes of the system; the

transformed coordinates x̃ = Vx are also called the modal coordinates.

The eigenvector vi defines the shape of the i-th mode;

The modal coordinate x̃i scales the mode (e.g., at the initial condition);
The eigenvalue λi defines how the amplitude of the mode evolves over time.

1 As an exponential e λi t x0 for real λi

2 As an sinusoid with exponentially changing amplitude e σt sin(ωit + φ0 )x0 for

complex-conjugate λi = σi + jωi .

3 As polynomially-scaled versions of the above t p e λi t for repeated λi .

The amplitude of the i mode goes to zero as t increases if and only if

Re(λi ) < 0.

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 8 / 26

Example: Simple Pendulum

Recall the model for a simple pendulum, assuming no damping (and no

input/output for now):  
0 1
ẋ = x
−g /l 0

Eigenvalues are solutions of the characteristic equation:

 
λ −1
det(λI − A) = det = λ2 + g /l = 0,
g /l λ

i.e., r
g
λ1,2 = ±j .
l

Eigenvectors are obtained as solutions of (λI − A)v = 0, e.g.,

 
1
v1,2 =
±j gl
p

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 9 / 26

Pendulum mode shape

Assume x(0) = (1, 0), i.e., x̃(0) = (1/2, 1/2).

Im Im

after time t:
multiply by e λi t
Re Re

Combining the two modes, we get

x1 (t) = cos ωt,

x2 (t) = −ω sin ωt.

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 10 / 26

Cartesian vs. polar form of complex numbers

Complex numbers can be represented in basically two ways:

Cartesian form: z = a + jb
Polar form: z = |z|e j∠z
√
In the above formulas, |z| = a2 + b 2 , and ∠z = arctan(b/a)
(arctan understood as the four-quadrant version).

The Cartesian form makes addition easy:

(a1 + jb1 ) + (a2 + jb2 ) = a1 + a2 + j(b1 + b2 )

The polar form makes multiplication easy:

m1 e jφ1 · m2 e jφ2 = (m1 + m2 )e j(φ1 +φ2 )

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 11 / 26

General evolution for complex-conjugate modes

Im Im

after time t:
multiply by e λi t
Re Re

Combining the two modes, we get

x1 (t) = c1 e σt sin(ωt + φ1 ,

x2 (t) = c2 e σt sin(ωt + φ2 ).

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 12 / 26

Towards feedback control

So far we have looked at how a given system, represented as a state-space

model or as a transfer function, behaves given a certain input (and/or initial
condition).

Typically the system behavior may not be satisfactory (e.g., because it is

unstable, or too slow, or too fast, or it oscillates too much, etc.), and one
may want to change it. This can only be done by feedback control!

However, we need to understand to what extent we can change the system

behavior. More precisely, for control design, one must also understand

how the control input can affect the state of the system;

how the state of the system affects the output.

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 13 / 26

Controllability and Observability

An LTI system of the form ẋ = Ax + Bu is said to be controllable if for any

given initial state x(0) = xc there exists a control signal that takes the state
to the origin x(t) = 0 for some finite time t.

An LTI system of the form ẋ = Ax + Bu, y = Cx + Du is said to be

observable if any given initial condition x(0) = xo can be reconstructed
based on the knowledge of the input and output signal only, over a finite time
interval [0, t].

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 14 / 26

Intuition from the Modal Form

Recall that if the matrix A has a complete set of independent (right)

eigenvectors {v1 , . . . , vn }, with eigenvalues
 {λ1 , . .. , λn } it can be
diagonalized by the matrix T = v1 v2 . . . vn .

The transformed state in the diagonalized system is such that x = T x̃.

The transformed model is (Ã, B̃, C̃ , D̃) = (T −1 AT , T −1 B, CT , D).

Component-wise, the dynamics of each modal coordinate x̃i are given by

d
x̃i (t) = λi x̃i (t) + b̃i u(t), i = 1, . . . , n.
dt

The output is given by

y = c̃1 x̃1 (t) + c̃2 x̃2 (t) + . . . + c̃n x̃n (t) + Du(t).

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 15 / 26

Controllability/observability for diagonal systems

From the previous slide, we can deduce

The i-th coordinate can be controlled if and only if b̃i 6= 0.

The i-th coordinate appears in the output if and only if c̃i 6= 0.

Hence:

An LTI system in diagonal form is controllable if b̃i 6= 0, i = 1, . . . , n.

An LTI system in diagonal form is observable if c̃i 6= 0, i = 1, . . . , n.

An LTI system is stabilizable if all unstable modes are controllable.

An LTI system is detectable if all unstable modes are observable.

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 16 / 26

An example

   
1.618j 0.4472j
 −1.618j  −0.4472j 
à =  , B̃ =  
 0.618j   0.4472 
−0.618j 0.4472
 
C̃ = 0.276j −0.276j 0.724 0.724 , D̃ = [0].

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 17 / 26

An example

   
j 0.707j
 −j  −0.707j 
à =  , B̃ =  
 j   0 
−j 0
 
C̃ = 0 0 0.707 0.707 , D̃ = [0].

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 18 / 26

More general conditions
Consider the derivatives of the output:
y (0) = Cx(0), y 0 (0) = CAx(0), y 00 (0) = CA2 x(0), . . .
One can reconstruct x(0), i.e., the system is observable, as long as the
observability matrix  
C
 CA 
 
 CA2 
 
 ... 
CAn−1
has full rank.
Note that it is sufficient to compute the observability matrix only up to the
power n − 1 of A (Cayley-Hamilton theorem).
A similar condition can be obtained for controllability, with a somewhat more
complicated proof: a system is controllable if the controllability matrix
A2 B An−1 B
 
B AB ...
has full rank.

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 19 / 26

Pole placement — first order systems

Consider a first-order control system with dynamics ẋ = ax + bu, and assume

that we are not happy about its behavior (e.g., it is unstable since a > 0, or
maybe stable but “slow” because |a| is small).

Can we change the behavior by choosing u in a clever way?

Feedback control: choose u = −kx; then the dynamics would become

ẋ = (a − bk)x

As long as b 6= 0 (i.e., if the system is controllable), by choosing

k = (a − a∗ )/b, we can place the ”closed-loop” eigenvalue at a desired value
a∗ , or anywhere we want on the real axis!

This is the simplest example of a general technique called “pole placement”.

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 20 / 26

Example

Assume ẋ = 0.5x + u, i.e., the system is unstable with time constant 2.

We would like the system to be stable, with time constant 1/2.

Choose u = −(0.5 + 2)x = −2.5x; the closed loop will be ẋ = −2x as

desired.
open-loop

closed-loop
u t

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 21 / 26

Effect on feedback for closed-loop dynamics
If we have an open-loop LTI system
ẋ(t) = Ax(t) + Bu(t),
y (t) = Cx(t),
by choosing a linear feedback u = −ky = −kCx, we can transform it into
another, closed-lopp LTI system:
ẋ(t) = (A − BkC )x(t),
y (t) = Cx(t),
In general “negative feedback” (i.e., u = −ky ) has “stabilizing” effects, and
the bigger k, the faster the closed-loop system is, and the smaller the errors
are.
However this is not generally the case.
In the rest of the course, we will look at ways to analyze the behavior of the
closed-loop system, and choosing the feedback control law, without
necessarily lots of computation — but rather using primarily “graphical”
methods.

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 22 / 26

Examples

ẋ(t) = 0.5x(t) + u(t),

y (t) = x(t),
Root Locus
0.3

0.2
Imaginary Axis (seconds-1 )

0.1

-0.1

-0.2

-0.3
-5 -4 -3 -2 -1 0 1
Real Axis (seconds-1 )

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 23 / 26

Examples
   
11 0
ẋ(t) = x(t) + u(t),
−1
1 1.5
 
y (t) = 1.34 0.67 x(t),
Root Locus
2.5

1.5

1
Imaginary Axis (seconds-1 )

0.5

-0.5

-1

-1.5

-2

-2.5
-7 -6 -5 -4 -3 -2 -1 0 1 2
Real Axis (seconds-1 )

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 24 / 26

Examples
   
−1 1 0 0
ẋ(t) = −1 −1 1  x(t) + 0 u(t),
0 0 −2 5
 
y (t) = 1 0 0 x(t),
Root Locus
4

2
Imaginary Axis (seconds-1 )

-1

-2

-3

-4
-6 -5 -4 -3 -2 -1 0 1
Real Axis (seconds-1 )
E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 25 / 26
Today’s learning objectives

After today’s lecture, you should be able to:

Diagonalize a matrix using similarity transformations.

Describe the behavior of an LTI system in modal coordinates.

Understand concepts like controllability and observability.

Understand the basic effects of feedback control on the closed-loop dynamics.

E. Frazzoli (ETH) Lecture 4: Control Systems I 13/10/2017 26 / 26