svdcmp.c File Reference

#include <math.h>
#include "useful_functions.h"
#include "nrfct.h"
#include "mbs_message.h"

Macros
#define	NRANSI

Functions
void	svdcmp (double **a, int m, int n, double w[], double **v)
	computes the Singular Value Decomposition a = u.w.v' More...
void	svdcmp_0 (double **a, int m, int n, double w[], double **v)
	computes the Singular Value Decomposition a = u.w.v' More...
double	pythag (double a, double b)

Variables
static double	dmaxarg1
static double	dmaxarg2
static int	iminarg1
static int	iminarg2
static double	dsqrarg

Detailed Description

This c file implements the functions to do the singular value decomposition of matrices.

Creation date: 2008

Author

Robotran team

Original file by JF Collard, 2008

(c) Universite catholique de Louvain

Macro Definition Documentation

NRANSI

#define NRANSI

Function Documentation

pythag()

double pythag	(	double	a,
		double	b
	)

svdcmp()

void svdcmp	(	double **	a,
		int	m,
		int	n,
		double	w[],
		double **	v
	)

computes the Singular Value Decomposition a = u.w.v'

The output matrix u is stored in the input matrix a.

The given arrays size is the minimum size, bigger arrays are acceptable. Indeed, the function will only use the first N elements of a 1D array or the first N rows and M columns of 2D arrays. Giving an oversized array as input (for example one of mbs_aux->... that is initialized at the system maximum possible size) is not an issue.

Parameters

[in,out]	a	the input matrix, with index starting at 1, to be decomposed. The content is modified to be the output matrix u. Its size must at least be m x n (0 element not taken into account).
[in]	m	the number of rows (0 element not taken into account) of the matrix a.
[in]	n	the number of columns (0 element not taken into account) of the matrix a.
[out]	w	the vector, with index starting at 1, to store the singular values.
[out]	v	the matrix v (not the transpose) of size n x n (0 element not taken into account).

svdcmp_0()

void svdcmp_0	(	double **	a,
		int	m,
		int	n,
		double	w[],
		double **	v
	)

computes the Singular Value Decomposition a = u.w.v'

The output matrix u is stored in the input matrix a. The given arrays size are minimum size, bigger arrays are acceptable.

Parameters

[in,out]	a	the input matrix, with index starting at 0, to be decomposed. The content is modified to be the output matrix u. Its size must be m x n.
[in]	m	the number of rows of the matrix a.
[in]	n	the number of columns of the matrix a.
[out]	w	the vector, with index starting at 0, to store the singular values.
[out]	v	the matrix v (not the transpose) of size n x n.

Variable Documentation

dmaxarg1

double dmaxarg1

static

dmaxarg2

double dmaxarg2

static

dsqrarg

double dsqrarg

static

iminarg1

int iminarg1

static

iminarg2

int iminarg2

static