《蚁群算法原理及其应用》（段海滨）附录Matlab源程序

2019年3月22日 245次阅读 来源: 蚁群算法

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% Function: Open and import file of city coordinates in TSP %%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

[fname,pname,filterindex]=uigetfile(‘*.*’,’All Files (*.*)’);
set(handles.text_filename,’String’,filename);
fid=fopen(filename,’r’);
if fid==-1
warndlg(‘Can not open the file’,’WARN’);
fclose(fid);
end
[citymat,COUNT]=fscanf(fid,’ %g’);
cN=COUNT/3;
set(handles.edit_citysum,’String’,cN);
fclose(fid);
aN=str2double(get(handles.edit_antsum,’String’));
ncmax=str2double(get(handles.edit_ncmax,’String’));
for nc=1:ncmax
for k=1:aN
g(nc,k)=fix((aN-1)*rand(1))+1;
end
end

% Basic Ant Colony Algorithm for TSP

global citymat;
global cN; % city number
global g

aN=str2double(get(handles.edit_antsum,’String’)); % Set ant number by using Matlab GUI
initao=str2double(get(handles.edit_tao,’String’));
alpha=str2double(get(handles.edit_alpha,’String’));
beta=str2double(get(handles.edit_beta,’String’));
Q=str2double(get(handles.edit_q,’String’));
rou=str2double(get(handles.edit_rou,’String’));
ncmax=str2double(get(handles.edit_ncmax,’String’));

%
% 原文中以上参数是通过作者自建的GUI界面读入的
% 为便于调用，可把程序改为Matlab的M函数文件，以上参数作为函数的输入
%
% function bestroute = fun_basic_TSP(aN,initao,alpha,beta,Q,rou,ncmax)
%


cities=reshape(citymat,3,cN); % get the city coordinates in TSP
x=cities(2,:);
y=cities(3,:);

for i=1:cN
for j=1:cN
dist(i,j)=sqrt((x(i)-x(j))^2+(y(i)-y(j))^2);
end
end

for i=1:cN
for j=1:cN
if j~=i
tao(i,j)=initao;
yita(i,j)=1/dist(i,j);
end
end
end

initao=initao.*ones(cN,cN);
detatao=zeros(cN,cN);
bestsolu=1e13;

%% This is the phase in which ants build their tours

for nc=1:ncmax
tabu=zeros(aN,cN);
for k=1:aN
tabu(k,g(nc,k))=1;
maxp(1)=0;
for j=1:cN
if tabu(k,j)==0
psum_medium0(1,j)=(tao(g(nc,k),j)^alpha).*(yita(g(nc,k),j)^beta);
else
psum_medium0(1,j)=0;
end
end
psum_medium=psum_medium0.’;
psum(k,1)=sum(psum_medium(:,1));
for j=1:cN
if tabu(k,j)==0
p(1,j)=(tao(g(nc,k),j)^alpha).*(yita(g(nc,k),j)^beta)/psum(k,1);
else
p(1,j)=0; % maybe it should be p(1,j)=0
end
if maxp(1)<p(1,j)
maxp(1)=p(1,j);
end
end
lindex=find(maxp(1)==p(1,:));
size1=[1,n];
[Rind,Cind]=ind2sub(size1,lindex(1));
solu_medium(k,1)=dist(g(nc,k),Cind(1));
route(k,1)=Cind(1);
tabu(k,Cind(1))=1;
tao(g(nc,k),Cind(1))=(1-rou)*tao(g(nc,k),Cind(1))+rou*initao(g(nc,k),Cind(1)); % local updating for the first iteration
tao(Cind(1),g(nc,k))=(1-rou)*tao(Cind(1),g(nc,k))+rou*initao(Cind(1),g(nc,k));
solu(k)=solu_medium(k,1);

for s=2:(cN-1)
Cind(s)=0;
Rind(s)=0;
lindex(s)=0;
maxp(s)=0;
for j=1:cN
if tabu(k,j)==0
psum_medium0(s,j)=(tao(route(k,s-1),j)^alpha).*(yita(route(k,s-1),j)^beta);
else
psum_medium0(s,j)=0;
end
end
psum_medium=psum_medium0.’;
psum(k,s)=sum(psum_medium(:,s));
for j=1:cN
if tabu(k,j)==0
p(s,j)=(tao(route(k,s-1),j)^alpha).*(yita(route(k,s-1),j)^beta)/psum(k,s);
else
p(s:(cN-1),j)=0;
end
if maxp(s)<p(s,j)
maxp(s)=p(s,j);
end
end
lindex=find(maxp(s)==p);
size2=[cN-1,cN];
[Rind(s),Cind(s)]=ind2sub(size2,lindex(1));
solu_medium(k,s)=dist(Cind(s-1),Cind(s));
route(k,s)=Cind(s);
tabu(k,Cind(s))=1;
tao(Cind(s-1),Cind(s))=(1-rou)*tao(Cind(s-1),Cind(s))+rou*initao(Cind(s-1),Cind(s)); % local updating for the first iteration
tao(Cind(s),Cind(s-1))=(1-rou)*tao(Cind(s),Cind(s-1))+rou*initao(Cind(s),Cind(s-1));
solu(k)=solu(k)+solu_medium(k,s);
end
tao(Cind(cN-1),g(nc,k))=(1-rou)*tao(Cind(cN-1),g(nc,k))+rou.*initao(Cind(cN-1),g(nc,k));
tao(g(nc,k),Cind(cN-1))=(1-rou)*tao(g(nc,k),Cind(cN-1))+rou.*initao(g(nc,k),Cind(cN-1)); % local updating for the first iteration
solu_medium(k,cN)=dist(Cind(cN-1),g(nc,k));
solu(k)=solu(k)+solu_medium(k,cN);
soluNC(nc,k)=solu(k);
bestsoluNC(nc)=soluNC(nc,1);
end
for k=1:m
if bestsoluNC(nc)>soluNC(nc,k)
bestsoluNC(nc)=soluNC(nc,k);
end
end

%% In this phase global updating occurs

lindex1=find(bestsoluNC(nc)==soluNC(nc,:));
size3=[nc,aN];
[Rind1(nc),Cind1(nc)]=ind2sub(size3,lindex1(1));
bestrouteNC(nc,:)=route(Cind1(nc),:);
[m,n]=size(lindex1);
for i=1:n
[Rind1t(nc,i),Cind1t(nc,i)]=ind2sub(size3,lindex1(i));

detatao(g(nc,Cind1t(nc,i)),route(Cind1t(nc,i),1))=Q/solu(Cind1t(nc,i));
detatao(route(Cind1t(nc,i),1),g(nc,Cind1t(nc,i)))=Q/solu(Cind1t(nc,i));
tao(g(nc,Cind1t(nc,i)),route(Cind1t(nc,i),1))=…
(1-rou).*tao(g(nc,Cind1t(nc,i)),route(Cind1t(nc,i),1))+…
rou.*detatao(g(nc,Cind1t(nc,i)),route(Cind1t(nc,i),1));
tao(route(Cind1t(nc,i),1),g(nc,Cind1t(nc,i)))=…
(1-rou).*tao(route(Cind1t(nc,i),1),g(nc,Cind1t(nc,i)))+…
rou.*detatao(route(Cind1t(nc,i),1),g(nc,Cind1t(nc,i)));

detatao(route(Cind1t(nc,i),n-1),g(nc,Cind1t(nc,i)))=Q/solu(Cind1t(nc,i));
detatao(g(nc,Cind1t(nc,i)),route(Cind1t(nc,i),n-1))=Q/solu(Cind1t(nc,i));
tao(route(Cind1t(nc,i),n-1),g(nc,Cind1t(nc,i)))=…
(1-rou).*tao(route(Cind1t(nc,i),n-1),g(nc,Cind1t(nc,i)))+…
rou.*detatao(route(Cind1t(nc,i),n-1),g(nc,Cind1t(nc,i)));
tao(g(nc,Cind1t(nc,i)),route(Cind1t(nc,i),n-1))=…
(1-rou).*tao(g(nc,Cind1t(nc,i)),route(Cind1t(nc,i),n-1))+…
rou.*detatao(g(nc,Cind1t(nc,i)),route(Cind1t(nc,i),n-1));

for s=2:n-1
detatao(route(Cind1t(nc,i),s),route(Cind1t(nc,i),s-1))=Q/solu(Cind1t(nc,i));
detatao(route(Cind1t(nc,i),s-1),route(Cind1t(nc,i),s))=Q/solu(Cind1t(nc,i));
tao(route(Cind1t(nc,i),s),route(Cind1t(nc,i),s-1))=…
(1-rou).*tao(route(Cind1t(nc,i),s),route(Cind1t(nc,i),s-1))+…
rou.*detatao(route(Cind1t(nc,i),s),route(Cind1t(nc,i),s-1));
tao(route(Cind1t(nc,i),s-1),route(Cind1t(nc,i),n-1))=…
(1-rou).*tao(route(Cind1t(nc,i),s-1),route(Cind1t(nc,i),s))+…
rou.*detatao(route(Cind1t(nc,i),s-1),route(Cind1t(nc,i),s));
end
end
if bestsolu>bestsoluNC(nc)
bestsolu=bestsoluNC(nc);
end

lindex2=find(bestsolu==bestsoluNC);
size4=[1,nc];
[Rind2,Cind2]=ind2sub(size4,lindex2(1));
bestroute(1,:)=bestrouteNC(Cind2,:);
initao=tao;
end
%% Output the best result for TSP

for nc=1:ncmax
bestrouteNC(nc,cN)=g(nc,Cind1(nc));
t(nc)=nc;
end
bestroute(1,cN)=bestrouteNC(Cind2,cN);
plot(x(bestroute(cN)),y(bestroute(cN)),’*’);
hold on;
for i=1:cN
plot(x(i),y(i),’o’);
hold on;
end
hold on;
line([x(bestroute(cN)) x(bestroute(1))],[y(bestroute(cN)) y(bestroute(1))]);
hold on;
for j=1:cN-1
line([x(bestroute(j)) x(bestroute(j+1))],[y(bestroute(j)) y(bestroute(j+1))]);
hold on;
end
hold off;
xlabel(‘X coordinate’);ylabel(‘Y coordinate’);
title(‘Best tour in NCmax iterations’);


    原文作者：蚁群算法
    原文地址: https://blog.csdn.net/jiav_net/article/details/83237433
    本文转自网络文章，转载此文章仅为分享知识，如有侵权，请联系博主进行删除。
点赞