Genetic algorithm initial popuplation

I use GA to minimize an objective function with 20 variables. I noticed that ga always produces initial population with small distance between individuals, usually at [0,1] while lower bound is 0 and upper bound is 90. I would like ga to produce the initial population more uniformly. Is this possible? My code:

n=20;
A=zeros(n,n);  %linear constraint x1<=x2<=x3<=x4<=...<=x20
b=zeros(n,1);

for i = 1:n
    for j = 1:n
        if(i+j==n+1)
            A(i,j)=-1;
        end
        if(i+j==n)
            A(i,j)=1;
        end
    end
end
b(n)=90;

LB=zeros(n,1);
UB=90*ones(n,1);

options=gaoptimset('creationfcn',@gacreationuniform,'PlotFcns',{@gaplotbestf,@gaplotscores,@gaplotbestindiv,@gaplotdistance},'Display','iter');

[x,fval,exitflag,output,final_pop]=ga(@(x) obj(x),n,A,b,[],[],LB,UB,[],options)

Answer 1

[Matt J]

use gaoptimset to control the Population options as you please

http://www.mathworks.com/help/gads/genetic-algorithm-options.html#f14223

Answer 2

[Alan Weiss]

I believe you need to set your PopInitRange option. See this example.

Alan Weiss

MATLAB mathematical toolbox documentation