Scribd
Upload
35 views

Controlability & Reachability

The system is not reachable in general as the controllability matrix is rank-deficient. Two of the eigenvalues (-1 and 1) are controllable while the third (-3) is uncontrollable. The system is stabilizable as the unstable eigenvalue is controllable. The set of reachable states is defined by a linear combination of the first two columns of the controllability matrix, representing a subspace rather than the entire state space.

Uploaded by

Rehman Saleem
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
35 views

Controlability & Reachability

The system is not reachable in general as the controllability matrix is rank-deficient. Two of the eigenvalues (-1 and 1) are controllable while the third (-3) is uncontrollable. The system is stabilizable as the unstable eigenvalue is controllable. The set of reachable states is defined by a linear combination of the first two columns of the controllability matrix, representing a subspace rather than the entire state space.

Uploaded by

Rehman Saleem
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Task 6: Controllability and Reachability [12 Points]

Consider the continuous-time system


   
−1 2 −1 −1
ẋ(t) =  0 1 1  x(t) +  1  u(t).
0 0 −3 0

a) Is the system reachable (in general)? Give reasons for your statement.

b) Identify the controllable and uncontrollable eigenvalues of the system using the
Popov-Belevitch-Hautus/rank test.

c) Is the system stabilizable? Give reasons for your statement.

d) Specify the set of all reachable states in form of a mathematical expression.

17
Solution to Task 6
a) [3 Points]
For LTI systems in continuous-time a completely state (or general) reachable system is also
completely state (or general) controllable and vice versa. As the controllability matrix C is
rank-deficient, the system is not reachable in general:
   
−1 2 −1 1 0 6
A =  0 1 1  , A2 = 0 1 −2 ,
0 0 −3 0 0 9
     
−1 3 −1
2
B = 1 , AB = 1 , A B = 1  ,
    
0 0 0

 
−1 3 −1
C = B, AB, A2 B =  1 1 1  ,
 

0 0 0

rank(C) = 2 < 3, det(C) = 0.

b) [4 Points]
Find the eigenvalues:
 
λ + 1 −2 +1
!
det([λI − A]) = det  0 λ − 1 −1  = (λ + 1) · (λ − 1) · (λ + 3) = 0,
0 0 λ+3

λ1 = −1,
λ2 = +1,
λ3 = −3.

Test the rank for each eigenvalue:


 
0 −2 1 −1
rank([λ1 I − A, B]) = rank 0 −2 −1 1  = 3,
0 0 2 0
 
2 −2 1 −1
rank([λ2 I − A, B]) = rank 0 0 −1 1  = 3,
0 0 4 0
 
−2 −2 1 −1
rank([λ3 I − A, B]) = rank  0 −4 −1 1  = 2 < 3.
0 0 0 0

The eigenvalues λ1 = −1 and λ2 = +1 are controllable while the eigenvalue λ3 = −3 is not.

c) [2 Points]
As the unstable eigenvalue (λ2 = +1, pos. real part) is controllable, the system is stabilizable.

18
d) [3 Points]
As the controllability matrix C is rank-deficient, the set of reachable or controllable states
xreachable is only a subspace. The subspace is defined by a linear combination of the fist linear
independent column vectors of C. In this case, only two out of three column vectors are linear
independent:

xreachable = α0 · B + α1 · AB
   
−1 3
= α0 ·  1  + α1 · 1 ,
0 0

with αi ∈ R.

19

You might also like