Remove close value for matrix

I have a matrix A with has dimension n x3

I want to remove the data values that are so close to certain value or repeated.

Let say A = [ 1 1 1; 1 1 1.5; 1.1 1.2 1 , so on ]

definitely they are closed and I cant use "unique" command because they are not repeated for all 3 values.

How to remove those value in certain distance: let say if the distance is less than or equal to .5, then remove that row. We can average those value that would be awesome all close value and replace it in the new matrix.

Note: A has unknown row, so it has to be a loop.

Answer 1

[Azzi Abdelmalek]

A = [ 1 1 1; 1 1 1.5; 1.1 1.2 1 ;2 3 4;2.1 2.9 4.5]
tol=0.5*ones(1,size(A,2));
v=(1:size(A,1));
idx=arrayfun(@(x) all(bsxfun(@le,abs(bsxfun(@minus,A,A(x,:))),tol),2),v,'un',0);
idx=cell2mat(idx);
for k=1:size(idx,1)
  q=idx(:,k);
  A(q,:)=repmat(mean(A(q,:)),sum(q),1);
end
out=unique(A,'rows')

Answer 2

[Matt J]

This one seems to be pretty popular

http://www.mathworks.com/matlabcentral/fileexchange/8354-consolidator

I'm sure you could browse the FEX for still more clustering algorithms.