Struct arcos_kdl::svd_eigen::SvdResult

pub struct SvdResult {
    pub u_matrix: DMatrix<f64>,
    pub s_vector: DVector<f64>,
    pub v_matrix: DMatrix<f64>,
    // some fields omitted
}

Struct for storing SVD results.

This struct Implements the Special algorithm for SVD used in Weighted damped least square kinematics.

Fields

u_matrix: DMatrix<f64>

Matrix holding left singular vectors

s_vector: DVector<f64>

Vector holding singular values

v_matrix: DMatrix<f64>

Matrix holding right singular vectors

Methods

impl SvdResult

pub fn new(num_joints: usize) -> Self

Create a default empty SvdResult

pub fn copy_jacobian_to_u_matrix(&mut self, jacobian: &Jacobian)

Copy a Jacobian to the matrix holding the left singular vectors

pub fn householder_reduction_to_bidiagonal(&mut self, epsilon: f64) -> f64

Reduction to bidiagonal form using the Householder Rotations method

pub fn accumulate_right_hand_transformations(&mut self, epsilon: f64)

Store the right singular vectors in its matrix (before convergence iterations)

pub fn accumulate_left_hand_transformations(&mut self, epsilon: f64)

Store the left singular vectors in its matrix (before convergence iterations)

pub fn diagonalize_bidiagonal_form(
    &mut self,
    maxiter: usize,
    anorm: f64,
    epsilon: f64
)

Find a diagonal matrix with the final singular values using an iterative Algorithm. Any changes in singular values order or magnitud is also reflected in singular vectors

pub fn compute(&mut self, jacobian: &Jacobian, epsilon: f64, maxiter: usize)

SVD computation with damped least squares using huseholder rotations

Trait Implementations

impl Clone for SvdResult

fn clone(&self) -> SvdResult

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.