Open zsiki opened 5 years ago
Revise reggeo.tcl to use tcllib where possible and Jacobi proc and GaussElimination proc.
Hi @zsiki .
SVD is optional. A completely sane solution is obtained with the simplest normalization of variables (Xn = X - M(X)
). For a detailed solution with Gaussian normalization (Xg = (X - M(X)) * sG; sG = 1 / sqrt(Tr(D(X)))
) see https://github.com/Geo-Linux-Calculations/geodesy-msexcel-ols/blob/main/spreadsheets/conformaltrans/helmert3d.xls . The condition number (cond(G)
) in this case turns out to be quite sane (in the example it is equal 38) and such a matrix can be solved by the usual LL decomposition.
for 3D coordinate transformation use tcllib::math module. ::math::linearalgebra::determineSVD