how should vector p in function place(a,b,p) be ?

pole placement algorithm is used in civil structural control. vector p should be computed based on open loop poles which are derived by p0=eig(a) . always eig(a) yields a vector of complex poles in descending order . does function place(a,b,p) expect vector p to be in descending order in order to compute gain matrix f (f=place(a,b,p)) correctly ?

Answer 1

[Azzi Abdelmalek]

The order of your poles vector doesn't matter

[jaden]

thank you for your answer . personally i think you are right because i found nothing in coding of place that would indicate any check on the order of poles by program . but i made an experiment out of this and assigned 2 different p vector one in ascending order and other in descending order . the results were close but NOT the SAME .

[Azzi Abdelmalek]

This is probably due to numerical errors