Controllability and Observability

Controllability and observability are two important features of linear

dynamic systems. If a systems is controllable, its closed-loop poles can be
assigned by state feedback, while if a system is observable, its state can be
reconstructed by an observer.

Definitions
Controllability: A system is completely controllable if each initial state
x ( t0 ) can be transferred to any final state x ( t f ) in finite time, by an unstrained
control u(t).
Observability: A system is completely observable if every initial state
x ( t0 ) can be uniquely determined from the output y(t) over the finite
interval of time t 0  t  t f .

In the simple words, controllable means every state can be driven by the
input, while observable means every state affects output.
 State Controllability
• A system is completely controllable if there exists an unconstrained
control u(t) that can transfer any initial state x(to) to any other
desired location x(t) in a finite time, to ≤ t ≤ T.

uncontrollable

controllable

If the inputs, or control forces, cannot in any way influence a particular state
variable then the system is said to be uncontrollable (not totally controllable –
but it may be partially controllable).
 State Controllability
Example 9:

Solution

The second equation does not share any elements with the first
equation and, in particular, does not contain an input term. Hence the
second state variable is not affected by any input which may be
applied and so it is uncontrollable.
 State Controllability
Example 10:

Solution
 State Controllability
• Controllability Matrix CM

CM  B  AB A2 B  An 1 B 
• System is said to be state controllable if

rank (CM )  n

The significance of this is that, since we have shown that the system is
controllable, the poles can be placed using state-variable feedback
design. If we had shown that the system was uncontrollable then there
would have been no point in continuing with a controller design.
Rank
Note that if an n*n matrix is nonsingular (that is, the matrix is of rank
n or the determinant is nonzero) then n column (or row) vectors are
linearly independent.

If the n*n matrix is singular (that is, the rank of the matrix is less
than n or the determinant is zero), then n column (or row) vectors are
linearly dependent.
 State Controllability
Example 11:
• Consider the system given below

 1 0  1
x    x   u
 0  3  0
y  1 2x
Solution
• State diagram of the system is U (s) 1 s-1 x1
1
Y (s)
1
2
3 2
x
s-1
 State Controllability
• Controllability matrix CM is obtained as
CM  B AB 
1  1
B  AB   
0  0

• Thus

1  1
CM   
0 0 
• Since ( )≠ therefore system is not completely
state controllable.
 Controllability and Observability
Example 12:
Evaluate controllability of a system with

0 6 0 
A  ,B   
Solution:
  1  5 1 
0 6 
Mc  B AB    
 1  5 
0 6
Mc   0  6  6  0
1 5
So, the system is completely state controllable.
 State Controllability
Example 13:

Solution
 Home Work

• Check the state controllability, state observability

and output controllability of the following system

0 1  0 
A  , B   , C  0 1
1 0 1
 State Observability
• A system is completely observable if and only if there exists a finite
time T such that the initial state x(0) can be determined from the
observation history y(t) given the control u(t), 0≤ t ≤ T.

observable
unobservable
 State Observability
• Observable Matrix (OM)
 C 
 CA 
 
Observability Matrix OM   CA 
2

 
  
CAn 1 

• The system is said to be completely state observable if

rank (OM )  n
 State Observability (Example)
• Consider the system given below
0 1  0 
x    x   u
0  2 1
y  0 4x

• OM is obtained as
C 
OM   
CA

• Where C  0 4

0 1 
CA  0 4   0  12
0  2 
 State Observability (Example)
• Therefore OM is given as
0 4 
OM   
 0  12 
• Since ( )≠ therefore system is not completely state
observable.

U(s) 1 -1
s x2 s-1 x1 Y(s)

2
 Controllability and Observability
Home Work

1- Examine controllability of the system with

0 1 0 0 
A  0 0 1  , B  0 
   
0  1  2 1

2- Examine observability of the system with

0 1 0
A  0 0 1 , C  1 0 0
 
0  1  2
 Controllability and Observability

3- Calculate controllability matrices and observability matrices for the

systems in previous Examples using Matlab functions, “ctrb” and “obsv”.
Then, evaluate controllability and observability of these systems using
Matlab function “rank”.