You are on page 1of 116

Example System: Predator Prey Model

https://www.math.duke.edu/education/webfeatsII/Word2HTML/Predator-prey.doc
Example System: Predator Prey Model
x  ax  bxy x y
y  cy  pxy

Without prey the predators


become extinct x  x  0
y  cy

Without predators the prey


Populations Oscillate grows unbounded x  ax
Two equilibrium: y y0
c a
(0,0) (x= ,y= ) (Chose a=100,b=1,c=100,p=1 to fix equilibrium at (100,100))
p b
Example System: Predator Prey Model
Linearization at the equilibrium points

e (  )  0

(100,100)
(0,0)
Chapter 3
Lyapunov Stability – Autonomous Systems

Question: Is this system “well behaved”?


System of Follow Up Question: What does well behaved mean?
differential • Do the state go to fixed values?
equations • Do the states stay bounded ?
• Do we know limits on the size of the states?

If we could “solve” the system then


theses questions may be easy to answer.
• We are assuming we can’t solve our
systems of interest.

Bottom line in this chapter is that we want to know if a differential equation,


which we can’t solve, is “well behaved”?
No explicit time dependence.
The solution will evolve with time, i.e. x(t)

Open or closed-loop system

Three main issues:


• System is nonlinear
• f is a vector
• Can’t find a solution

Can have multiple equilibrium points

We will work to quantify “well behaved”


Evolution of the state
Khalil calls this the “Challenge and answer
form to demonstrate stability”
• Challenger proposes an ε bound for the
final state
• The answerer has to produce a  bound on
the initial condition so that the state always
stays in the ε bound.
• Answerer has to provide an answer for
every ε proposed

Bottom line: If we start close


enough to xe we stay close to xe
Pendulum without friction.
Is (0,0) a stable equilibrium point in the sense of Definition 2?

Yes
 
Initial condition is  , 0 
x1 ' = x2 a = 10  6 
x2 ' = - a sin(x1)

1
x2

0 x1 ' = x2 a = 10
x2 ' = - a sin(x1)
-1
0.6

-2 0.4

0.2
-3
x1 0
-4
-0.2
-4 -3 -2 -1 0 1 2 3 4
x1 -0.4

Give me an  -0.6
We can find a  -6 -4 -2 0 2 4 6
t
x(t) goes to xe as t goes to 

A equilibrium point could be Stable and not Convergent


An equilibrium point could be Convergent and Not Stable
Pendulum without friction.
Is (0,0) convergent? No
Is (0,0) asymptotically stable? No

x1 ' = x2 a = 10
x2 ' = - a sin(x1)

1
x2

0 x1 ' = x2 a = 10
x2 ' = - a sin(x1)
-1
0.6

-2 0.4

0.2
-3
x1 0
-4
-0.2
-4 -3 -2 -1 0 1 2 3 4
x1 -0.4

-0.6

-6 -4 -2 0 2 4 6
t
Pendulum with friction.
Yes
Is (0,0) a stable equilibrium point in the sense of Definition 2?
Is (0,0) convergent? Yes
Is (0,0) asymptotically stable? Yes
x1 ' = x2 a = 10
x2 ' = - a sin(x1) - 5 x2

1
x2

-1
x1 ' = x2 a = 10
x2 ' = - a sin(x1) - 5 x2
-2
0.6

-3 0.5

0.4
-4

x1
0.3
-4 -3 -2 -1 0 1 2 3 4
x1 0.2

0.1

0 0.5 1 1.5 2 2.5 3


t
Includes a sense of “how fast” the system converges.

Exponential is “stricter” than asymptotic


Can perform this shift to any equilibrium point of interest
(a system may have multiple equilibrium points)
V is a scalar function

Notation:
PSD write as V0
PD write as V>0
NSD write as V0
ND write as V<0

Could be V(x1(t)) Example: Are each of these PSD, PD, ND, or NSD?
V(x1) V(x1) V(x1) V(x1)
None

ND

x1 x1 x1 x1
PSD PD
Could be x1(t)
x =x T x

V(x1) "if and only if"


PD if q11  0
Example:
Example: V(x)=x T Qx where x 
then Q  (i.e. a scalar)
PSD or NSD if q11  0 we can write a general example as V(x)=q11x12
x1

ND q11  0
Substitute original system
x  f ( x)

x 
V is a general function
x   1 of x1 and x2
 x2 

V(x) is specific to this


system of interest
Don’t loose generality by restricting Q to be symmetric in quadratic form

i.e., Q symmetric

See chapter 2

I.e. only the symmetric part


a b  
T a 0  0 b   contributes to the quadratic
V  xT   x  x     x
c d    0 d   c 0 
 ax12  dx22
Where is this going?
x12 could be an abstraction of the
energy stored in the system
• Analogy would work for potential
V(x)
energy of a spring or charge on a
PD V>0 capacitor

Let’s say that x1 is the


x1
state of our system

V(x1(t)) Now if we find: V(x(t))  0


How could that happen?
V(x1(t)) will
not increase Could be
negative

t
Can take as many derivatives as
you need

Eventually we will perform the


control design to make this true
Br is a ball about the origin. Our
Theorem only applies at the origin Every convergent
sequence has a
convergent
subsequence in Br
V ( x)  0

Closeness of x
to x=0 implies
closeness of
V(x) to V(0)
xe  0 V (0)  0

Definition of
stability
q
l

mg

Not required to be able to


derive these equations for
this class.
Example 3 (cont)

l-h q
l

v
lh
cos(q )  h
l
l  h  l cos(q ) mg
h  l 1  cos(q ) 
Example 3 (cont)

Note: open interval that


does not include 2, 2

Note:
•Stable implies that the system (pendulum position and velocity) remains bounded
•Does not mean that the system has stopped moving
•The energy is constant (V=E) but the system continuously moves exchanging kinetic and
potential energy Limit Cycle)
Limit Cycles
Oscillation leads to a closed path in the phase plane, called periodic orbits. For
example, the pendulum has a continuum of closed paths
If there is a single, isolated periodic orbit such that all trajectories tend to that
periodic orbit then it is called a stable limit cycle.

Can also have unstable limit cycles.

Khalil, Nonlinear Systems 3rd Ed, p59


Example: LC Oscillator

dv di 1
iC V L  i   vdt
dt dt L

Kirchoff's current law at node 1


dv 1

into node
i  C
dt
  L
vdt  0

Differentiate, define x1  v, x2  v
x1  x2
1
x2   x1
LC

Khalil, Nonlinear Systems 3rd Ed, p59


Example: Oscillator

iR1 iR2

v=h(i)

Khalil, Nonlinear Systems 3rd Ed, p59


Example: Oscillator

x1  v
x2  x1  v
friction
We would expect that the
friction will take energy out
of the system, thus the
system will stop moving.

ND? Physically, we know that the system will stop because of


the friction but our analysis does not show this (we only
NSD? show it is Stable but can’t show Asymptotic Stability).
We will fix this problem using the Invariant Set Theorem
Example 4 (cont)

We know the system is asymptotically stable even if


that is not illustrated by this specific Lyapunov
analysis.

Note: We will return to this later and fix using the Invariant Set
Theorem.
This is a Local stability result because
we have limited the range of the state
variables for which the result applies.
Have we learned anything useful?
Eq point
0    1  1.1 x  x  0
Given control design problem:

x  ax  bu x  , constants a, b  0
input designed to stabilize system as u  k p x
x  ax  bk p x    a  bk p  x  f ( x )

What values (if any) of k p will work? Eq point


0    1  1 x  x  anything

V  x2
V  2 xx  2 x (a  bk p ) x  2(ax  k p ) x 2
a
V  0 if 2( a  bk p )  0  k p 
b

Given the actual system a=1, b=2: Eq point


0    1  0.9  x  x  0
1
kp 
2
There is a missing piece in our original proof that prevents us from applying the result
globablly:

V  0 (PD) and V  0 (ND)  AS at x e


that is, V ( x)  V ( x(0))
V ( x) decreasing

Problem: V ( x) might be decreasing (as to the Theorem)


but V ( x) may not be sufficient
to capture the behavior of the states (globablly)
x
Evolution of the state
(green line)

State x1 gets large but it is


trapped by the constant
Lyapunov function

Constant V(x1,x2) contours

Not radially unbounded because the x1 term in V  1 as x1  


i.e. small V does not imply small x1
Add this condition to AS Theorem:

Fixes the problem that the


Theorem 2 was local.

Global AS: The states will go to zero as time increases from any finite starting state
Now want to address the problem demonstrated in the pendulum with friction example,
couldn’t show function was negative definite using energy arguments.

scalar function

like radially
unbounded
V is PD iif there are class K functions that upper and lower bound the Lyapunov function.
Raleigh-Ritz Theorem.

class   functions
Summary: If the state starts within some ball, then the state remains within some other ball
for all times.

We will eventually use t here


Reminder of Theorem 2 (need on next slide)
V (0)  0
V ( x)  0
V ( x)  0

Solve rhs and substitute


a new upper bound
Solve lhs V p
  x
K2
Solve the Substitute a new upper bound
differential
inequality
V ( x0 ) p
 x0
K2
x(t ) is less than an exponential bound p
“What happens in M stays in M”

Now want to address the problem demonstrated in the pendulum with friction example,
couldn’t show function was negative definite using energy arguments.
An oscillator has a limit cycle

Slotine and Li, Applied Nonlinear Control


This is a condition
that we can try to test
The system x  f ( x ) defines the trajectory
 Use knowledge of the trajectory to test this condition
The only place when V ( x)  0 is when the system is at the equilibrium
iv)

The system x  f ( x) defines the trajectory


"vanish identically" -> exactly =0
(vs. approaching zero)
i) PD iv) Radially Unbounded

ii) NSD

Stays at x2=0 forever,


i.e. doesn’t move
trajectory

iii) Doesn’t vanish along a trajectory except x1=0


Different from Lyapunov Theorem because:
1) V does not need to be Positive Definite
2) Applies to multiple equilibrium points and limit cycles
Example of LaSalle’s Theorem
System:
a,  are positive constants
Equilibrium Points:

Suggests the
Lyapunov Function Candidate: possibility of adding  in V
1 2 1 2 1 2 1 2
V  x1  x2 V  x1  x2
2 2
2 2
V  x1 x1  x2 x2   x2 x12  ax12   x12 x2
  1    x2 x12  ax12
V  0 (PD),V  0 ??? (no)
Can we change the It would be
Lyapunov function to yield a nice of this
better derivative? was a “1”
Example of LaSalle’s Theorem (cont)
Lyapunov Function Candidate:

1 2 1 2
Can we change the V  x1  x2
Lyapunov function to 2 2
yield a better
derivative?
V  x1 x1  x2 x2   x2 x12  ax12   x12 x2
  1    x2 x12  ax12
V  0 (PD),V  0 ??? (no)
Would be nice if we could turn into  bx12
where the  bx12 is more negative than ax12 is positive
Suggests the
possibility of changing V further
1
V  ...   x2  b 
2

2
1 1 1
V  ...   x2  b  x2  ..  x2 x2  bx2
  
V  ..  x2 x12  bx12
Example of LaSalle’s Theorem (cont)

"Checklist"
We have enough
information to Set D  2
conclude the system is
“stable” in some region
i ) M  D invariant wrt system
-> set is invariant wrt ii ) V  0 in M
the system
iii ) E where V  0
iv ) N largest invariant set in E
Example of LaSalle’s Theorem (cont)
Lyapunov Function Candidate:

Conclusion from Theorem

"Checklist"
x2 (t )  constant as t   Set D  2
i ) M  D invariant wrt system
ii ) V  0 in M
iii ) E where V  0
iv ) N largest invariant set in E
?
x1 ' = 3 x2
x2 ' = - 5 x1 + x13 - 2 x2 D
4

1
x2

-1 However, it does look like


there should be a region such
-2
that the system converges.
-3 Can we define that region?
-4

-4 -3 -2 -1 0 1 2 3 4
x1
Define Region of Attraction (RA):
We are assuming we know
there is a bound on x2

Sub
Original assumption , V is
Sub
decreasing
Sub
k 0 Solve differential
Sub inequality

k
Find RA for k=2:
2
k  x(0) system is ES
RA for the specific case of k  2 :
2
x(0)  2
x(0)  2
x ' = - k x + x3 k=1

 2  x(0)  2
y '=0
This is our RA:
1

0.8

0.6
k=2
Phase portrait:
0.4

0.2

0
y

-0.2 1 0 x
 2 2
-0.4
Estimate Region of Attraction (RA) Using LaSalle’s Theorem:
Finish Example 14
Finish Example 14 (cont)
V(x)
150

100

50

0
4
2 4
0 2 x1 ' = 3 x2
x2 0 x2 ' = - 5 x1 + x13 - 2 x2
-2
-2
-4 -4 x1 4

Contour lines of V(x) x2


1

-1

-2

-3

-4

-4 -3 -2 -1 0 1 2 3 4
x1
Proven the “If” part of the Theorem. If Q>0 and found P
then AS real part of eigenvalues is >0

Now prove the “only If” part of the Theorem.


Example: Is the following system AS?
solve 3 equations with 3 unknowns

 x1  0 1  x1  0 2 0   p1  1 
 x   1 1  x  1 1 1  p   0 
 2   2   2  
0 2 2   p3  1 
p p2   x1 
V   x1 x2   1  p1   1.5 
 p2 p3   x2   p    0.5
 2  
Q  ( AT P  PA)  p3   1 
Choose an arbitrary Q that is PD and symetric  1.5 0.5
P
1 0   0.5 1 
easy pick Q  I 2 x 2   
0 1  1  0.6910, 2  1.8090
1 0    0 1   p1 p2   p1 p2  0 1  Found a PD, symmetric P from an arbitrary Q
   p p p 
0 1 
    1 1 2 3  2 p3  1 1   eigenvalues of A satify e(i )  0
 the origin of the system is AS
1 0    p2 p3   p2  p1  p2  
0 1        p p  p 
   1 p  p2  p2  p 3  3 2 3 In MATLAB: >> P=lyap([0 -1;1 -1],[1 0;0 1])

1 0   2 p2 p3  p1  p2  
0 1      Wouldn’t it be easier just to find the
    p 3  p1  p2  2 p2  2 p3  eigenvalues of A? Yes, but having P (and
solve 3 equations with 3 unknowns hence a Lyapunov function) will be useful
later.
Linearization of a Nonlinear System

= 0 by assumption of equilibrium point Assume small

Change of variables to make the origin the equilibrium


Example: Linearize System

 x22  x1 cos( x2 ) 
f  

 2 1 1
x x  x  1   x1 sin( x )
2 
f  cos( x2 ) 2 x2  x1 sin( x2 ) 
 
x  2 x1  1  sin( x2 ) 1  x1 cos( x2 ) 
f 1 0 

x x [0,0]T 1 1 

Eigenvalues = 1,1

Origin is unstable
Example : Linearize System (cont)
Original System
x1 ' = x22 + x1 cos(x2)
x2 ' = x2 + (x1 + 1) x1 + x1 sin(x1) x1 ' = x1
Linearized
x2 ' = x1 + x2

0.1
0.1
0.08
0.08
0.06
0.06

0.04
0.04

0.02
0.02
x2

x2
0

-0.02
-0.02

-0.04
-0.04

-0.06
Compare favorably -0.06

-0.08
close to the equilibrium
-0.08

-0.1 -0.1

-0.1 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 point


0.08 0.1 -0.1 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 0.1
x1 x1

x1 ' = x22 + x1 cos(x2) x1 ' = x1


x2 ' = x2 + (x1 + 1) x1 + x1 sin(x1) x2 ' = x1 + x2

2
2

1
1

0
0
x2

-1
x2

-1

-2
-2

-3
May compare less
favorably further away
-3

-4
from equilibrium point -4
-2 -1 0 1 2 3 4
-2 -1 0 1 2 3 4
x1
x1
xe 

x0
In general, proving a system is
unstable is not very
“constructive” in designing
control systems.
Unstable equilibrium
V  ()(  2  x )
2

V  0 if x  
Example

x1  x2
x2  x2  x1
1 2

V  x1  x22
2

V  2 x1 x2  x22

Can we define a set U  V>0 ?


Doesn't appear possible because in region about the origin
x0 
would have to include a point x1  0 and x 2 =0
which would make V<0
Not conclusive, doesn’t mean it is stable
Summary
x=0 stable

convergent

x=0 asymptotically stable

x=0 exponentially stable

x=0 stable  convergent x=0 asymptotically stable  x=0 exponentially stable

X(t) X(t) X(t) X(t)

time time time time


10
Summary
Local, global
x=0 asymptotically stable

2 3
1

x=0 stable
x=0 asymptotically stable
Locally, NSD dV/dt x=0 asymptotically stable
Locally, state could grow
while V shrinks globally
V is class K
8
7

x=0 exponentially stable


x=0 asymptotically stable local global, local depend on conditions

9 Conditions true in entire state space

Global, asymptotically stable


Summary

Linearization

11

x=0 exponentially stable


Inherently local result since it is an approximation of a nonlinear system at a point

x1 ' = x22 + x1 cos(x2)


x2 ' = x2 + (x1 + 1) x1 + x1 sin(x1) x1 ' = x1
x2 ' = x1 + x2
2
2

1
1

0
0
x2

-1
x2

-1

-2
-2

-3

system -3
Linear approximation
-4
-4
-2 -1 0 1 2 3 4
x1 -2 -1 0 1 2 3 4
x1
Summary
u
f(x)

x  f ( x)  u
if we use feedback of the state i.e. u( x )  g ( x ) then

the analysis tools in this


chapter will be the basis
u
f(x) for designing the
control u(x)
f2(x)
g(x)

x  f2 ( x)
is still autonomous
Example System: Predator Prey Model

https://www.math.duke.edu/education/webfeatsII/Word2HTML/Predator-prey.doc
Example System: Predator Prey Model
x  ax  bxy x y
y  cy  pxy

Without prey the predators


become extinct x  x  0
y  cy

Without predators the prey


Populations Oscillate grows unbounded x  ax
Two equilibrium: y y0
c a
(0,0) (x= ,y= ) (Chose a=100,b=1,c=100,p=1 to fix equilibrium at (100,100))
p b
Example System: Predator Prey Model
Linearization at the equilibrium points

e (  )  0

(100,100)
(0,0)
Homework #3-1
• Find all of the equilibrium points for the
pendulum without friction.
– Plot the phase portrait from -3<x1<3 and -
10<x2<10.
– From inspection of the phase portrait , does it
appear that the equilibrium points are stable?
– Use the Lyapunov function candidate from the
notes to examine stability at x1= and 2
0  x2
Homework (sol)
0  a sin( x1 )
Equilibrium points at (0,0), (  n ,0) for n  1, 2,3...
Pendulum spins

Pendulum swings

Pendulum
doesn’t move
(stable EQ point)

Pendulum doesn’t
move at exactly that
point
(Unstable EQ point)
Is EQ at x1   stable? Homework (sol)
Change of variable Shifted system
y1  x1   then y1  x1 y1  y2
y2  x2 then y2  x2 g
y2   sin( y1   )
l
1 2 2
V ml y2  mgl (1  cos( y1 )) Same Lyapunov function candidate that we used in notes
2
y
y y y y

 V V   f1 ( y ) 
V ( y)   
y2   f 2 ( y ) 
,
 y1 V ( y)  0 ? No
 y2 
  mgl sin( y1 ) , ml y2   g
2  Is the EQ point No
  sin( y1   )  stable or conclusion
 l 
unstable? can be made
 mgly2 sin( y1 )  mgly2 sin( y1   )
 2mgly2 sin( y1 )
Is EQ at x1  2 stable? Homework (sol)
Change of variable Shifted system
y1  x1  2 then y1  x1 y1  y2
y2  x2 then y2  x2 g
y2   sin( y1  2 )
1 2 2 l
V ml y2  mgl (1  cos( y1 )) Same Lyapunov function candidate that we used in notes
2
y
y y y y

 V V   f1 ( y ) 
V ( y)    V ( y)  0 ?
y2   f 2 ( y ) 
, Yes
 y1
 y2 
  mgl sin( y1 ) , ml y2   g
2  Is the EQ point
  sin( y1  2 )  stable or
 l  Stable
unstable?
 mgly2 sin( y1 )  mgly2 sin( y1  2 )
0
Homework #3-2
Find the equilibrium points for each of the following and use a quadratic
Lyapunov function candidate to determine stability of each equilibrium point.

1) x   x 3) x1  x2 5) x1  x2  x3  x1
2) x   x  5 x2   x1  x2 x2   x1  x3  x2
4) x1   x2 x3   x1  x2  x3

x2  x1  1  x12 x2 
Using a quadratic Lyapunov function, find u=f(x) such that the system is stable at
x=0 6) x  x  u
specify u  f ( x)
Using a quadratic Lyapunov function, find the conditions on a,b,c,d such that the
system is stable at x=0
7) x1  ax1  bx2
x2  cx1  dx2
Homework (sol)
1) x   x
Eq pt at x  0
1 2
V x
2
V  xx  x   x    x 2
V  0 (PD), V  0 (ND)  AS (at x=0)
Homework (sol)
2) x   x  5
Eq pt at x  5
Define change of variables y  x  5 then y  x
New system y   y
1 2
V y
2
V  yy  y   y    y 2
V  0 (PD), V  0 (ND)  AS at y=0  AS at x=-5
Homework (sol)
3) x1  x2
x2   x1  x2
Eq pt at x1  x2  0
1 1 0   x1  1 2 1 2
V  [ x1 x2 ]      x1  x2
2 0 1   x2  2 2
V  x1 x1  x2 x2  x1  x2   x2   x1  x2 
  x2 2
V  0 (PD), V  0 (NSD)  Stable
Homework (sol)
4) 4) x1   x2

x2  x1  1  x12 x2 
Eq pt at x1  x2  0
1 2 1 2
V  x1  x2
2 2
 
V  x1 x1  x2 x2  x1   x2   x2 x1  1  x12 x2  
  x1 x2  x1 x2  x2 2  x12 x2 2   x2 2 1  x 
2
1

V  0 (PD), V  0 (NSD)  Stable ISL


Homework (sol)
5) x1  x2  x3  x1
x2   x1  x3  x2
x3   x1  x2  x3
Eq pt at x1  x2  x3  0
1 2 1 2 1 2
V  x1  x2  x3
2 2 2
V  x1 x1  x2 x2  x3 x3  x1  x2  x3  x1   x2   x1  x3  x2   x3 ( x1  x2  x3 )
  x12  x2 2  x32
V  0 (PD), V  0 (ND)  AS
Homework (sol)
6) x  x  u
Assume Eq pt at x  0
1 2
V x
2
V  xx  x 2  xu
would like to get rid of the unhelpful x 2
and add a stabilizing term  x 2
let u  -2 x (or even u  - Kx with K>1)
V  x2  2x2   x2
V  0 (PD), V  0 (ND)  AS
Note that x   x has an EQ point at zero as we assumed.
We just designed a feedback control to stabilize the system
We used the Lyapunov function to design the control.
7) x1  ax1  bx2 Homework (sol)
x2  cx1  dx2
Assume Eq pt at x1  x2  0
1 1
V  x12  x2 2
2 2
V  x1 x1  x2 x2
 ax12  bx1 x2  cx1 x2  dx2 2
 ax12  dx2 2   b  c  x1 x2
choose a and d to stabilize the system ie a  0 and d  0
choose b and c to make the last term "go away" ie b  c  0
then
V  0 (PD), V  0 (NSD)  Stable

Since this is a linear system, compare to finding conditions for positive eigenvalues:
 a b   0 
A  I      0     a    d     bc
 c d   
 2  (a  d )  ad  bc  0
Homework 3.3
Chapter 3 - Problems 3.1, 3.3, 3.6, 3.7
Homework 3.3 (sol)

System doesn’t move, i.e. derivatives are zero

x1  0  x2
 
x1  0   x1  x13  x2  x1 1  x12  0  x1  0,1, 1
equilibrium points at (0,0), (1,0), and (-1,0)

Notation means the second eq point, not the square of the eq point

See that (0,0) which corresponds to (1,0) in the original


system is an equilibrium point
Homework 3.3 (sol)
(-1,0)

See that (0,0) which corresponds to (-1,0) in the original


system is an equilibrium point
Homework 3.3 (sol)

Solve for conditions on v (note that this is the voltage on the coil not a
Lyapunov function candidate) so that x1=yo and the system is at rest, i.e. solve
v so that there is an equilibrium point at x1=yo. Approach: set derivatives to
zero, x1 to the contant yo solve for v.
Emphasizing that we
are looking for a
Homework 3.3 (sol) constant x3

x1=yo
Homework 3.3 (sol)

This is why
linearization can be so
powerful tool - you
get to use all of the
linear analysis tools.
Homework 3.3 (sol)

a) Verify that the origin is an equilibrium


Substitute (0,0)
x1  0
x2  0
thus (0,0) is an equilibrium point
Homework 3.3 (sol)
Homework 3.3 (sol)
From pplane8:  0

 1   1
Homework 3.3 (sol) Test this first

V can be zero other than


at the origin
Homework 3.3 (sol)
Homework 3.4
Chapter 3 - Problems 3.4,3.5,3.13, 3.14
Homework 3.4 (Sol)
Homework 3.4 (Sol)

For small x2
Because we
assumed small x2
Homework 3.4 (Sol)
Homework 3.4 (Sol)

a) Substitute (0,0) in the system, then x1  x2  0


 the system can't move from the point
 x1  x2  0 for all time which means in stays in the ste (0,0).
 The equilibrium point is an invariant set.
Note that any equilibrium point is an invariant set.

b) The set defines a relationship between x1 and x 2 as: 1  (3 x12  2 x22 )  0  3 x12  1  2 x22  x1 
1
3
 
1  2 x22 ;

i.e. points (
1
3
 
1  2 x22 , x2 )

d
Invariance implies that there is no deviation from the set over time, ie ( set )  0
dt
 x 
d
 
1  (3 x12  2 x22 ) 
x
 
1  (3 x12  2 x22 ) x   6 x1 4 x2   1   6 x1 x1  4 x2 x2
dt  x2 
2 
3 
 
Sub from system: =  6 x1  x2   4 x2  x1  x2 1  3 x12  2 x22    4 x 1  3x
2
2
2
1  2 x22 
Sub from our set definition (using 3 x12  1  2 x22 ): =0
Thus the set is invariant w.r.t the system.
Homework 3.4 (Sol)

This is a clever way to add a degree of freedom to a


quadratic Lyapunov function. a,b don’t change the
basic nature of the quadratic function, ie still PD and
radially unbounded.

a,b provide the opportunity to cancel these


cross terms which might have other wise
stopped our analysis (ie had we chosen a=b=1)
Homework 3.4 (Sol)
Why choose this? Because someone tried a lot
of other V’s that did not work.

You might also like