How can I integrate into a fitness function another code.

am trying to use the GA on my existing project(simulator). So I have created the algorithm and I am trying to integrate it into my existing code. So what I am trying to do is instead of using the "y" function, I am trying to use my project code. But it seems that it does not like it when I am commenting out the Y, or even when I am using y=@Project_code;

clear all;
 
car_range_max=[98,52];     
car_range_min=[-98,-52];   
 
opts=gaoptimset('PlotFcns',@gaplotbestf,'PopulationSize',10,'PopInitRange',[car_range_max;car_range_min],'Generations',20,'CrossoverFraction',0.7,'TolFun',0.0001,'MutationFcn',@mutationadaptfeasible);  

global z
z = [];
nvars = 2;
[bestx, minfval_M]=ga(@untitled2,nvars,[],[],[],[],car_range_max,car_range_min,[],opts);  
 
bestx   
fval 
%/*********************** another file ***************************************/
function y= untitled2(x)
global z
max_speed=x(1);                                   
acc_speed=x(2);                                  
 
[y,car_energy,car_travel_time]= Project_code(max_speed,acc_speed); % Project_code is expecting the values from max_speed and acc_speed
 
z=[z;max_speed,acc_speed,car_energy,car_travel_time];

end
%Bellow is redundant as I will use the function Project_code and not the y
%funtion
%function [y, car_energy,car_travel_time] = Project_code(max_speed, acc_speed)
   % y = max_speed+acc_speed+4*max_speed+4*acc_speed+8;
   % max_speed = randi(10); acc_speed = randi(10);
%end

Answer 1

[Walter Roberson]

global z
z = [];

nvars = 2;
[bestx, fval] = ga(@untitled2, nvars)

Optimization terminated: maximum number of generations exceeded.

bestx = 1×2

-2.0049 -1.9989

fval = 2.5557e-05

z

z = 9453×4

5.5635 -2.5558 7.0000 6.0000 9.7908 1.5695 2.0000 8.0000 4.6551 0.7655 10.0000 10.0000 -1.8040 -9.4902 8.0000 3.0000 0.4168 -9.8843 6.0000 10.0000 0.1650 5.2319 9.0000 3.0000 -1.0891 2.2510 6.0000 5.0000 -8.3005 6.3059 5.0000 8.0000 -9.9609 -1.9774 1.0000 2.0000 2.5086 9.1399 9.0000 9.0000

function y  = untitled2(x)
    global z
    max_speed=x(1);                                      
    acc_speed=x(2);                                    
    [y, a, b] = Project_code(max_speed, acc_speed);
    
    z=[z;max_speed,acc_speed,a,b]; % store the data into a table
end

function [y, a, b] = Project_code(max_speed, acc_speed)
    y = max_speed^2+acc_speed^2+4*max_speed+4*acc_speed+8;
    a = randi(10); b = randi(10);
end

[I Bos]

Thanks very much for your reply.

it shows me this error Execution of script Project_code as a function is not supported:
I have edited my first post in order to be more clear.

Caused by:
    Failure in initial user-supplied fitness function evaluation. GA cannot continue.

[Walter Roberson]

One thing I notice is that your car_range_max is being passed in the lb (lower bound) position, and your car_range_min is being passed in the ub (upper bound) position. But your max is positive and your min is negative, so you are saying that the lower bound is positive and the upper bound is negative. You need to fix that.

Meanwhile:

what result do you get if you call

untitled2([0, 0])

[I Bos]

when I call that, it shows me the below:

Execution of script Project_code as a function is not supported:

/MATLAB Drive/Project_code.m

Error in untitled2 (line 6)

[y,car_energy,car_travel_time]= Project_code(max_speed,acc_speed);

[Walter Roberson]

Well that's a problem.

If Project_Code is currently a script that expects to receive values in certain variables and make assignments to particular variables... then it is really recommended that you convert it to a be a function. Calling a script from a function can lead to subtle problems.

[I Bos]

Yes, you are correct, the Project_Code expect some values ie. max_speed and acc_speed. When I was trying to call that script Project_Code. I was geeting the result with some errors as well.