Find minimum of function using genetic algorithm in Simulink

Hi

Thank you for reading this question!

I try to apply the solver "ga" in Simulink. Then, the simulation shows errors, which is "Function 'ga' not supported for code generation". After, I added the command "coder.extrinsic('ga')" in front of the code. However, the error is "Function handles cannot be passed to extrinsic functions." The code and simulation are shown below. I'm not sure if the solver "ga" can be applied to Simulink. Could anyone help me or share the relevant link?

Many thanks in advance!

function [y, fval, exitflag] = fcn(lb, ub)

coder.extrinsic('ga')

fun = @(x)100*(x(2)-x(1)^2)^2 + (1-x(1))^2;

A = [];

b = [];

Aeq = [];

beq = [];

[y, fval, exitflag] = ga(fun,2,A,b,Aeq,beq,lb,ub)

end

Answer 1

[Ayush Aniket]

Hi xin lin,

As the error suggests, you need to refactor your code so that you don't pass function handles across the extrinsic function call boundary. You can wrap up all of that code in yet another function, let's call it myCode.m. Then, declare that whole function as extrinsic and call it from your MATLAB Function block as shown below:

function [y, fval, exitflag] = fcn(lb, ub)
    coder.extrinsic('mycode');
    y = zeros(1,2);%preintialize this with expected dimensions
    fval = zeros(1);%preintialize this with expected dimensions
    exitflag = zeros(1);%preintialize this with expected dimensions
    [y, fval, exitflag] = myCode(lb,ub);
end

function [y, fval, exitflag] = myCode(lb,ub) %define this in MATLAB workspace
    fun = @(x)100*(x(2)-x(1)^2)^2 + (1-x(1))^2;
    A = [];
    b = [];
    Aeq = [];
    beq = [];
    [y, fval, exitflag] = ga(fun,2,A,b,Aeq,beq,lb,ub);
end

Hope this helps!

[xin lin]

Hi Ayush,

Thanks for answer. This is very helpful.

Best regards,

Xin