%
% This function compute a one d.o.f. controller for
% a n-th order SISO STRICTLY proper plant (continuous or discrete)
% described in state space form by the 4-tuple (a,b,c,0)
%
%
% The user must supply a vector po with n desired
% observer poles and a vector pc with n desired
% control poles
%
% The function returns the gains K and J and
% the numerator and denominator polynomial of the controller
% transfer function (see Chapter 8 in
% "Principles of Control System Design"
% by Goodwin, Graebe and Salgado
%
%
% Warning: This function uses MATLAB function "place", thus the
% user is advised to become familiar with the numerical
% limitations of the algorithm used in "place".
%
[n,n]=size(a);
%
[n1,n2]=size(po);
if max(n1,n2)~=n
error ('illegal number of observer poles')
end
[n3,n4]=size(pc);
if max(n3,n4)~=n
error ('illegal number of controller poles')
end
%
A=real(a-J*c-b*K); B=real(J);C=real(K);
%
% Controller reduction
%
[Am,Bm,Cm,Dm]=minreal(A,B,C,0);
[num,denc]=ss2tf(Am,Bm,Cm,Dm);
[nx,numc]=p_elcero(num);
%
