Search code examples
matlaboptimizationleast-squares

Fsolve with variables,

I have this matlab function,

function [f]=ErrorFun(a,b,c)
    global I     
    global phi

    f = sum((a+b.*cos(phi)+c.*sin(phi)-I).^2);
end

length(a)=length(b)=length(phi)=length(c)=length(I)N.I want to use fsolve, but I don't know how to do it. ErrorFun is a least squares problem.

Solution

  • The documentation says

    X = fsolve(FUN,X0) starts at the matrix X0 and tries to solve the equations in FUN. FUN accepts input X and returns a vector (matrix) of equation values F evaluated at X.

    so you need to rewrite your function so that it accepts a single vector of parameters as its input. E.g., in your case, 

    function f = ErrorFun(x)
    global I phi
    n = length(phi) ;
    a = x(1:n) ;
    b = x(n+1:2*n) ;
    c = x(2*n+1:3*n) ;

    f = sum((a+b.*cos(phi)+c.*sin(phi)-I).^2) ;
end

    then call fsolve with this function and some initial vector x0 = [a; b; c].