Home > 04-Estimation > GHM > Functions-GHM > cg_2D_line.m

cg_2D_line

PURPOSE ^

% determines constraint value and Jacobians for 2D line estimation

SYNOPSIS ^

function [cg,A,B] = cg_2D_line(lv,lnu,xa)

DESCRIPTION ^

% determines constraint value and Jacobians for 2D line estimation

 g= ^l2 - a ^l1 -b = 0  (== y_i-ax_i-b=0,i=1,...,N)

 lv     = 2*G vector of observations [x;y]
 lnu    = 2*G vector of approximate fitted observations [x^a;y^a]
 xa     = 2x1 vector, approximate values for parameters xa=[a;b]

 cg     = Gx1 vector of residual of constraints
 A      = GxU Jacobian dg/dx
 B      = NxG Jacobian (dg/dl)'

 Wolfgang Förstner
 wfoerstn@uni-bonn.de

CROSS-REFERENCE INFORMATION ^

This function calls: This function is called by:

SOURCE CODE ^

0001 %% determines constraint value and Jacobians for 2D line estimation
0002 %
0003 % g= ^l2 - a ^l1 -b = 0  (== y_i-ax_i-b=0,i=1,...,N)
0004 %
0005 % lv     = 2*G vector of observations [x;y]
0006 % lnu    = 2*G vector of approximate fitted observations [x^a;y^a]
0007 % xa     = 2x1 vector, approximate values for parameters xa=[a;b]
0008 %
0009 % cg     = Gx1 vector of residual of constraints
0010 % A      = GxU Jacobian dg/dx
0011 % B      = NxG Jacobian (dg/dl)'
0012 %
0013 % Wolfgang Förstner
0014 % wfoerstn@uni-bonn.de
0015 
0016 function  [cg,A,B] = cg_2D_line(lv,lnu,xa)
0017 
0018 % number of constraints
0019 G2 = size(lv,1);
0020 G  = G2/2;                        % == N
0021 
0022 % Jacobians
0023 A  = - [lv(1:G) ones(G,1)];       % G x U
0024 B = [-xa(1)*eye(G) eye(G)]';      % N x G
0025 
0026 % residual of constraints
0027 cg = -(lv(G+1:2*G) - xa(1) * lv(1:G) - xa(2)) + B'*(lnu-lv);  % G x 1
0028 
0029

Generated on Sat 21-Jul-2018 20:56:10 by m2html © 2005