A solver for mixed LCP problems using the Gauss-Seidel iterative method. More...

#include <SurgSim/Math/MlcpGaussSeidelSolver.h>

Inheritance diagram for SurgSim::Math::MlcpGaussSeidelSolver:

Public Member Functions
	MlcpGaussSeidelSolver ()
	Constructor. More...

	MlcpGaussSeidelSolver (double epsilonConvergence, double contactTolerance, size_t maxIterations)
	Constructor. More...

virtual	~MlcpGaussSeidelSolver ()
	Destructor. More...

bool	solve (const MlcpProblem &problem, MlcpSolution *solution)
	Resolution of a given MLCP (Gauss Seidel iterative solver) More...

double	getEpsilonConvergence () const

void	setEpsilonConvergence (double precision)
	Set the precision. More...

double	getContactTolerance () const

void	setContactTolerance (double tolerance)
	Set the contact tolerance. More...

size_t	getMaxIterations () const

void	setMaxIterations (size_t maxIterations)
	Set the max number of iterations. More...

Public Member Functions inherited from SurgSim::Math::MlcpSolver
	MlcpSolver ()
	Constructor. More...

virtual	~MlcpSolver ()

Private Member Functions
void	computeEnforcementSystem (size_t problemSize, const MlcpProblem::Matrix &A, const MlcpProblem::Vector &b, const MlcpSolution::Vector &initialGuessAndSolution, const std::vector< MlcpConstraintType > &constraintsType, size_t constraintID, size_t matrixEntryForConstraintID)

void	calculateConvergenceCriteria (size_t problemSize, const MlcpProblem::Matrix &A, const MlcpProblem::Vector &b, const MlcpSolution::Vector &initialGuessAndSolution, const std::vector< MlcpConstraintType > &constraintsType, double constraintConvergenceCriteria[MLCP_NUM_CONSTRAINT_TYPES], double convergenceCriteria, bool validSignorini)

void	doOneIteration (size_t problemSize, const MlcpProblem::Matrix &A, const MlcpProblem::Vector &b, MlcpSolution::Vector initialGuessAndSolution, const MlcpProblem::Vector &frictionCoefs, const std::vector< MlcpConstraintType > &constraintsType, double constraintConvergenceCriteria[MLCP_NUM_CONSTRAINT_TYPES], double convergenceCriteria, bool *validSignorini)

void	printViolationsAndConvergence (size_t problemSize, const MlcpProblem::Matrix &A, const MlcpProblem::Vector &b, const MlcpSolution::Vector &initialGuessAndSolution, const std::vector< MlcpConstraintType > &constraintsType, double convergenceCriteria, bool validSignorini, size_t iterations)

Private Attributes
double	m_epsilonConvergence
	The precision. More...

double	m_contactTolerance
	The contact tolerance. More...

size_t	m_maxIterations
	The maximum number of iterations. More...

size_t	m_numEnforcedAtomicConstraints
	The number of atomic constraints, aka the system size. More...

Matrix	m_lhsEnforcedLocalSystem
	The left-hand side matrix. More...

Vector	m_rhsEnforcedLocalSystem
	The right-hand side vector. More...

std::shared_ptr< SurgSim::Framework::Logger >	m_logger
	The logger. More...

Detailed Description

A solver for mixed LCP problems using the Gauss-Seidel iterative method.

The problem can contain:

CONSTRAINT = Bilateral constraint (all atomic, a fixed 3D point=3 atomics independents constraints)
CONTACT = Unilateral constraint
- frictionless => 1 atomic constraint per contact
- frictional with Coulomb friction (1 mu parameter per contact) => 3 atomic dependent constraints per contact (1 directional + 2 tangentials)
SUTURING = Sliding constraint for suturing
- Frictionless suturing constraint => 2 atomic constraints per sliding point
- Frictional suturing constraint => 3 atomic constraints per sliding point (2 directional + 1 tangential with friction on it) => 1 mu parameter per frictional suturing

See e.g.: Duriez, Christian; Dubois, F.; Kheddar, A.; Andriot, C., "Realistic haptic rendering of interacting deformable objects in virtual environments," IEEE Transactions on Visualization and Computer Graphics, vol.12, no.1, pp.36,47, Jan.-Feb. 2006.

Constructor & Destructor Documentation

◆ MlcpGaussSeidelSolver() [1/2]

SurgSim::Math::MlcpGaussSeidelSolver::MlcpGaussSeidelSolver ( )

Constructor.

◆ MlcpGaussSeidelSolver() [2/2]

SurgSim::Math::MlcpGaussSeidelSolver::MlcpGaussSeidelSolver	(	double	epsilonConvergence,
		double	contactTolerance,
		size_t	maxIterations
	)

Constructor.

Parameters

epsilonConvergence	The precision.
contactTolerance	The contact tolerance.
maxIterations	The max iterations.

◆ ~MlcpGaussSeidelSolver()

SurgSim::Math::MlcpGaussSeidelSolver::~MlcpGaussSeidelSolver ( )

virtual

Destructor.

Member Function Documentation

◆ calculateConvergenceCriteria()

void SurgSim::Math::MlcpGaussSeidelSolver::calculateConvergenceCriteria	(	size_t	problemSize,
		const MlcpProblem::Matrix &	A,
		const MlcpProblem::Vector &	b,
		const MlcpSolution::Vector &	initialGuessAndSolution,
		const std::vector< MlcpConstraintType > &	constraintsType,
		double	constraintConvergenceCriteria[MLCP_NUM_CONSTRAINT_TYPES],
		double *	convergenceCriteria,
		bool *	validSignorini
	)

private

◆ computeEnforcementSystem()

void SurgSim::Math::MlcpGaussSeidelSolver::computeEnforcementSystem	(	size_t	problemSize,
		const MlcpProblem::Matrix &	A,
		const MlcpProblem::Vector &	b,
		const MlcpSolution::Vector &	initialGuessAndSolution,
		const std::vector< MlcpConstraintType > &	constraintsType,
		size_t	constraintID,
		size_t	matrixEntryForConstraintID
	)

private

◆ doOneIteration()

void SurgSim::Math::MlcpGaussSeidelSolver::doOneIteration	(	size_t	problemSize,
		const MlcpProblem::Matrix &	A,
		const MlcpProblem::Vector &	b,
		MlcpSolution::Vector *	initialGuessAndSolution,
		const MlcpProblem::Vector &	frictionCoefs,
		const std::vector< MlcpConstraintType > &	constraintsType,
		double	constraintConvergenceCriteria[MLCP_NUM_CONSTRAINT_TYPES],
		double *	convergenceCriteria,
		bool *	validSignorini
	)

private

◆ getContactTolerance()

double SurgSim::Math::MlcpGaussSeidelSolver::getContactTolerance ( ) const

Returns: The contact tolerance.

◆ getEpsilonConvergence()

double SurgSim::Math::MlcpGaussSeidelSolver::getEpsilonConvergence ( ) const

Returns: The precision.

◆ getMaxIterations()

size_t SurgSim::Math::MlcpGaussSeidelSolver::getMaxIterations ( ) const

Returns: The max number of iterations.

◆ printViolationsAndConvergence()

void SurgSim::Math::MlcpGaussSeidelSolver::printViolationsAndConvergence	(	size_t	problemSize,
		const MlcpProblem::Matrix &	A,
		const MlcpProblem::Vector &	b,
		const MlcpSolution::Vector &	initialGuessAndSolution,
		const std::vector< MlcpConstraintType > &	constraintsType,
		double	convergenceCriteria,
		bool	validSignorini,
		size_t	iterations
	)

private

◆ setContactTolerance()

void SurgSim::Math::MlcpGaussSeidelSolver::setContactTolerance ( double tolerance )

Set the contact tolerance.

Parameters

tolerance The contact tolerance.

◆ setEpsilonConvergence()

void SurgSim::Math::MlcpGaussSeidelSolver::setEpsilonConvergence ( double precision )

Set the precision.

Parameters

precision The precision.

◆ setMaxIterations()

void SurgSim::Math::MlcpGaussSeidelSolver::setMaxIterations ( size_t maxIterations )

Set the max number of iterations.

Parameters

maxIterations The max number of iterations.

◆ solve()

bool SurgSim::Math::MlcpGaussSeidelSolver::solve	(	const MlcpProblem &	problem,
		MlcpSolution *	solution
	)

virtual

Resolution of a given MLCP (Gauss Seidel iterative solver)

Parameters

	problem	The mlcp problem
[out]	solution	The mlcp solution

Returns: true if successfully converged.

Implements SurgSim::Math::MlcpSolver.

Member Data Documentation

◆ m_contactTolerance

double SurgSim::Math::MlcpGaussSeidelSolver::m_contactTolerance

private

The contact tolerance.

◆ m_epsilonConvergence

double SurgSim::Math::MlcpGaussSeidelSolver::m_epsilonConvergence

private

The precision.

◆ m_lhsEnforcedLocalSystem

Matrix SurgSim::Math::MlcpGaussSeidelSolver::m_lhsEnforcedLocalSystem

private

The left-hand side matrix.

◆ m_logger

std::shared_ptr<SurgSim::Framework::Logger> SurgSim::Math::MlcpGaussSeidelSolver::m_logger

private

The logger.

◆ m_maxIterations

size_t SurgSim::Math::MlcpGaussSeidelSolver::m_maxIterations

private

The maximum number of iterations.

◆ m_numEnforcedAtomicConstraints

size_t SurgSim::Math::MlcpGaussSeidelSolver::m_numEnforcedAtomicConstraints

private

The number of atomic constraints, aka the system size.

◆ m_rhsEnforcedLocalSystem

Vector SurgSim::Math::MlcpGaussSeidelSolver::m_rhsEnforcedLocalSystem

private

The right-hand side vector.

The documentation for this class was generated from the following files:

SurgSim/Math/MlcpGaussSeidelSolver.h
SurgSim/Math/MlcpGaussSeidelSolver.cpp

Public Member Functions

Private Member Functions

Private Attributes

Detailed Description

Constructor & Destructor Documentation

◆ MlcpGaussSeidelSolver() [1/2]

◆ MlcpGaussSeidelSolver() [2/2]

◆ ~MlcpGaussSeidelSolver()

Member Function Documentation

◆ calculateConvergenceCriteria()

◆ computeEnforcementSystem()

◆ doOneIteration()

◆ getContactTolerance()

◆ getEpsilonConvergence()

◆ getMaxIterations()

◆ printViolationsAndConvergence()

◆ setContactTolerance()

◆ setEpsilonConvergence()

◆ setMaxIterations()

◆ solve()

Member Data Documentation

◆ m_contactTolerance

◆ m_epsilonConvergence

◆ m_lhsEnforcedLocalSystem

◆ m_logger

◆ m_maxIterations

◆ m_numEnforcedAtomicConstraints

◆ m_rhsEnforcedLocalSystem