ga custom functions parameters

Hi everyone,

I am trying to optimize a race track, represented by a class, with the ga function. Since I use a class and not some numbers to optimize, I need to write all the needed functions myself (mutation, crossover, etc). I have a really hard time figuring out what exactly the input parameters of the functions are and what they are supposed to return.

As an example: I first thought the fitness function gets one individual and returns the fitness value, but apparently it gets a 1x1 cell array with one individual. Or the crossover function: the 'parents' parameter is a vector with two integers. I assume those are the indices of the parents in the 'thisPopulation' parameter, but honestly I have no clue.

Can someone help me understand what happens here? Even just a short description of what goes in/out would help. Unfortunately the official documentation is not very helpful in terms of variable types. The functions I need are the CreationFcn, FitnessFcn, MutationFcn and CrossoverFcn. Here is my code at the moment.

Thanks for your efforts, the help is appreciated.

Answer 1

[Alan Weiss]

I am sorry that you are having trouble. Perhaps this example will help.

Alan Weiss

MATLAB mathematical toolbox documentation

[Theobald Beyer]

Thanks for the reply, unfortunately I already read this page and it doesn't help much. All the function parameters are explained, but only in a semantic way ( "PARENTS: Parents chosen by the selection function"), I'd also need the syntactic description: "PARENTS: 2x1 cell array of class x".