state-space controlability

hello.

i'm fairly new to MatLab and im trying calculate whether a state-space system is controllable or not and then display it in writing. the question i have is: am i doing it correclty? i've googled a bit trying to figure it out on my own, but since i have little no no experience using the software i'm not getting a definite answer.

my goal is simply to write out whether the system is controlable or not, and im unsure if i've got the syntax correct.

thanks in advance :)

System controlability.

clc, clear, close all;

A=[1 1;-4 -2;]; B=[1 -1;1 -1];

system=ss(A,B,[],[]);

co=ctrb(system)

dimensjon=isequal(A,co);

if(dimensjon==0)

disp('system is not controlable')

else

disp('system is controlable')

end

edit:

i've since updated the code:

• think it's better, but original question still stands.

System controlability.

clc, clear, close all;

A=[1 1;4 -2]; B=[1 -1;1 -1];

system=ss(A,B,[],[]);

co=ctrb(system);

unco= length(A)-rank(co);

if(unco==0)

disp('system is controlable')

else

disp('system is not controlable')

end

Answer 1

[Star Strider]

The example in the documentation section on Check System Controllability is the same test I usually use. It would be appropriate in your application as well.

[sverre Kvist]

thank you.

so if i continue with:

system=ss(A,B,[],[]);

co=ctrb(system);

unco= length(A)-rank(co);

if(unco==0)

disp('system is controlable')

else

disp('system is not controlable')

end

i could rely on the answer of the if-sentence. the way i understood it is that as long as unco is 0 it is equivalent to det(B AB) !=0.

[Star Strider]

It is important to remember that this function (and every other one that I am aware of) is subject to the constrints and inaccuracies of floating-point calculations. (See the documentation section on Limitations for details.) It may be necessary to use a tolerance in any test.

A=[1 1;-4 -2;]; B=[1 -1;1 -1];
system=ss(A,B,[],[]);
co=ctrb(system)

co = 2×4

1 -1 2 -2 1 -1 -6 6

unco = length(system.A) - rank(co)

unco = 0

I am not certain that calculating ‘det([system.B system.A*system.B])’ is even possible since:

Test = det([system.B system.A*system.B])

Error using det
Matrix must be square.

I would use the test in the documentation.

.

[sverre Kvist]

alright. thank you very much for your assistance.

i think i've arrived at a code that works.

for Reference:

clc, clear, close all;

A=[1 1;4 -2]; B=[1 -1;1 -1];

system=ss(A,B,[],[]);

controlabilityMatrix=ctrb(A,B); %Mc=[B, AB]

unco= length(A)-rank(controlabilityMatrix);

if(unco==0)

disp('system is controlable')

else

disp('system is not controlable')

end

[Star Strider]

My pleasure!

The code appears to be correct, at least wirh respect to my understanding.

If my Answer helped you solve your problem, please Accept it!

.