from
QCAT
by Ola Harkegard Quadratic Programming Control Allocation Toolbox
Description of wpinv
wpinv
PURPOSE
WPINV - Compute weighted pseudoinverse.
SYNOPSIS
function [G,F] = wpinv(A,W)
DESCRIPTION
WPINV - Compute weighted pseudoinverse.
[G,F] = wpinv(A,W)
Computes the optimal solution x = Gy + Fx0 to the least squares
problem
min ||W(x-x0)|| subj. to Ax = y
x
See also PINV.
CROSS-REFERENCE INFORMATION
This function calls:
This function is called by:
SOURCE CODE
0001 function [G,F] = wpinv(A,W)
0002
0003 % WPINV - Compute weighted pseudoinverse.
0004 %
0005 % [G,F] = wpinv(A,W)
0006 %
0007 % Computes the optimal solution x = Gy + Fx0 to the least squares
0008 % problem
0009 %
0010 % min ||W(x-x0)|| subj. to Ax = y
0011 % x
0012 %
0013 % See also PINV.
0014
0015 [m,n] = size(A);
0016 if nargin < 2
0017 W = eye(n);
0018 end
0019
0020 % Thesis, Lemma B.1
0021 G = inv(W)*pinv(A*inv(W));
0022 F = eye(n) - G*A;