Backward correction for optimal control problems of unconstrained rigid-body systems. More...

#include <unconstr_backward_correction.hpp>

Public Member Functions
	UnconstrBackwardCorrection (const OCP &ocp, const int nthreads)
	Construct a backward correction. More...

	UnconstrBackwardCorrection ()
	Default constructor. More...

	~UnconstrBackwardCorrection ()=default
	Destructor. More...

	UnconstrBackwardCorrection (const UnconstrBackwardCorrection &)=default
	Default copy constructor. More...

UnconstrBackwardCorrection &	operator= (const UnconstrBackwardCorrection &)=default
	Default copy operator. More...

	UnconstrBackwardCorrection (UnconstrBackwardCorrection &&) noexcept=default
	Default move constructor. More...

UnconstrBackwardCorrection &	operator= (UnconstrBackwardCorrection &&) noexcept=default
	Default move assign operator. More...

void	setNumThreads (const int nthreads)
	Sets the number of threads of the parallel computations. More...

void	initAuxMat (aligned_vector< Robot > &robots, const std::vector< GridInfo > &time_discretization, const Solution &s, KKTMatrix &kkt_matrix, KKTResidual &kkt_residual)
	Initializes the auxiliary matrices by the terminal cost Hessian computed by the current solution. More...

void	initConstraints (aligned_vector< Robot > &robots, const std::vector< GridInfo > &time_discretization, const Solution &s)
	Computes the cost and constraint violations. More...

void	evalOCP (aligned_vector< Robot > &robots, const std::vector< GridInfo > &time_discretization, const Eigen::VectorXd &q, const Eigen::VectorXd &v, const Solution &s, KKTResidual &kkt_residual)
	Computes the cost and constraint violations. More...

void	evalKKT (aligned_vector< Robot > &robots, const std::vector< GridInfo > &time_discretization, const Eigen::VectorXd &q, const Eigen::VectorXd &v, const Solution &s, KKTMatrix &kkt_matrix, KKTResidual &kkt_residual)
	Computes the KKT residual and matrix. More...

void	coarseUpdate (aligned_vector< Robot > &robots, const std::vector< GridInfo > &time_discretization, const Eigen::VectorXd &q, const Eigen::VectorXd &v, const Solution &s, KKTMatrix &kkt_matrix, KKTResidual &kkt_residual)
	Eval KKT and coarse updates the solution leveraging the parallel computation. More...

void	backwardCorrection (const std::vector< GridInfo > &time_discretization, const Solution &s, const KKTMatrix &kkt_matrix, const KKTResidual &kkt_residual, Direction &d)
	Performs the backward correction for coarse updated solution and computes the Newton direction. More...

double	primalStepSize () const
	Gets the maximum primal step size of the fraction-to-boundary-rule. More...

double	dualStepSize () const
	Gets the maximum dual step size of the fraction-to-boundary-rule. More...

const PerformanceIndex &	getEval () const
	Gets the performance index of the evaluation. More...

void	integrateSolution (const aligned_vector< Robot > &robots, const std::vector< GridInfo > &time_discretization, const double primal_step_size, const double dual_step_size, Direction &d, Solution &s)
	Integrates the solution. More...

void	integratePrimalSolution (const aligned_vector< Robot > &robots, const std::vector< GridInfo > &time_discretization, const double primal_step_size, const Direction &d, Solution &s)
	Integrates the solution. More...

Detailed Description

Backward correction for optimal control problems of unconstrained rigid-body systems.

Constructor & Destructor Documentation

◆ UnconstrBackwardCorrection() [1/4]

robotoc::UnconstrBackwardCorrection::UnconstrBackwardCorrection	(	const OCP &	ocp,
		const int	nthreads
	)

Construct a backward correction.

Parameters

[in]	ocp	Optimal control problem.
[in]	nthreads	Number of the threads of the parallel computations. Must be positive.

◆ UnconstrBackwardCorrection() [2/4]

robotoc::UnconstrBackwardCorrection::UnconstrBackwardCorrection ( )

Default constructor.

◆ ~UnconstrBackwardCorrection()

robotoc::UnconstrBackwardCorrection::~UnconstrBackwardCorrection ( )

default

Destructor.

◆ UnconstrBackwardCorrection() [3/4]

robotoc::UnconstrBackwardCorrection::UnconstrBackwardCorrection ( const UnconstrBackwardCorrection & )

default

Default copy constructor.

◆ UnconstrBackwardCorrection() [4/4]

robotoc::UnconstrBackwardCorrection::UnconstrBackwardCorrection ( UnconstrBackwardCorrection && )

defaultnoexcept

Default move constructor.

Member Function Documentation

◆ backwardCorrection()

void robotoc::UnconstrBackwardCorrection::backwardCorrection	(	const std::vector< GridInfo > &	time_discretization,
		const Solution &	s,
		const KKTMatrix &	kkt_matrix,
		const KKTResidual &	kkt_residual,
		Direction &	d
	)

Performs the backward correction for coarse updated solution and computes the Newton direction.

Parameters

[in]	time_discretization	Time discretization.
[in]	s	Solution.
[in,out]	kkt_matrix	KKT matrix.
[in,out]	kkt_residual	KKT residual.
[in,out]	d	Direction.

◆ coarseUpdate()

void robotoc::UnconstrBackwardCorrection::coarseUpdate	(	aligned_vector< Robot > &	robots,
		const std::vector< GridInfo > &	time_discretization,
		const Eigen::VectorXd &	q,
		const Eigen::VectorXd &	v,
		const Solution &	s,
		KKTMatrix &	kkt_matrix,
		KKTResidual &	kkt_residual
	)

Eval KKT and coarse updates the solution leveraging the parallel computation.

Parameters

[in,out]	robots	aligned_vector of Robot.
[in]	time_discretization	Time discretization.
[in]	q	Initial configuration.
[in]	v	Initial generalized velocity.
[in]	s	Solution.
[in,out]	kkt_matrix	KKT matrix.
[in,out]	kkt_residual	KKT residual.

◆ dualStepSize()

double robotoc::UnconstrBackwardCorrection::dualStepSize ( ) const

Gets the maximum dual step size of the fraction-to-boundary-rule.

Returns: The dual step size of the fraction-to-boundary-rule.

◆ evalKKT()

void robotoc::UnconstrBackwardCorrection::evalKKT	(	aligned_vector< Robot > &	robots,
		const std::vector< GridInfo > &	time_discretization,
		const Eigen::VectorXd &	q,
		const Eigen::VectorXd &	v,
		const Solution &	s,
		KKTMatrix &	kkt_matrix,
		KKTResidual &	kkt_residual
	)

Computes the KKT residual and matrix.

Parameters

[in,out]	robots	aligned_vector of Robot.
[in]	time_discretization	Time discretization.
[in]	q	Initial configuration.
[in]	v	Initial generalized velocity.
[in]	s	Solution.
[in,out]	kkt_matrix	KKT matrix.
[in,out]	kkt_residual	KKT residual.

◆ evalOCP()

void robotoc::UnconstrBackwardCorrection::evalOCP	(	aligned_vector< Robot > &	robots,
		const std::vector< GridInfo > &	time_discretization,
		const Eigen::VectorXd &	q,
		const Eigen::VectorXd &	v,
		const Solution &	s,
		KKTResidual &	kkt_residual
	)

Computes the cost and constraint violations.

Parameters

[in,out]	robots	aligned_vector of Robot.
[in]	time_discretization	Time discretization.
[in]	q	Initial configuration.
[in]	v	Initial generalized velocity.
[in]	s	Solution.
[in,out]	kkt_residual	KKT residual.

◆ getEval()

const PerformanceIndex & robotoc::UnconstrBackwardCorrection::getEval ( ) const

inline

Gets the performance index of the evaluation.

Returns: const reference to the performance index.

◆ initAuxMat()

void robotoc::UnconstrBackwardCorrection::initAuxMat	(	aligned_vector< Robot > &	robots,
		const std::vector< GridInfo > &	time_discretization,
		const Solution &	s,
		KKTMatrix &	kkt_matrix,
		KKTResidual &	kkt_residual
	)

Initializes the auxiliary matrices by the terminal cost Hessian computed by the current solution.

Parameters

[in]	robots	aligned_vector of Robot.
[in]	time_discretization	Time discretization.
[in]	s	Solution.
[in,out]	kkt_matrix	KKT matrix.
[in,out]	kkt_residual	KKT residual.

◆ initConstraints()

void robotoc::UnconstrBackwardCorrection::initConstraints	(	aligned_vector< Robot > &	robots,
		const std::vector< GridInfo > &	time_discretization,
		const Solution &	s
	)

Computes the cost and constraint violations.

Parameters

[in,out]	robots	aligned_vector of Robot for paralle computing.
[in]	time_discretization	Time discretization.
[in]	s	Solution.

◆ integratePrimalSolution()

void robotoc::UnconstrBackwardCorrection::integratePrimalSolution	(	const aligned_vector< Robot > &	robots,
		const std::vector< GridInfo > &	time_discretization,
		const double	primal_step_size,
		const Direction &	d,
		Solution &	s
	)

Integrates the solution.

Parameters

[in,out]	robots	aligned_vector of Robot for paralle computing.
[in]	time_discretization	Time discretization.
[in]	primal_step_size	Primal step size.
[in]	d	Direction.
[in,out]	s	Solution.

◆ integrateSolution()

void robotoc::UnconstrBackwardCorrection::integrateSolution	(	const aligned_vector< Robot > &	robots,
		const std::vector< GridInfo > &	time_discretization,
		const double	primal_step_size,
		const double	dual_step_size,
		Direction &	d,
		Solution &	s
	)

Integrates the solution.

Parameters

[in,out]	robots	aligned_vector of Robot for paralle computing.
[in]	time_discretization	Time discretization.
[in]	primal_step_size	Primal step size.
[in]	dual_step_size	Dual step size.
[in,out]	d	Direction.
[in,out]	s	Solution.

◆ operator=() [1/2]

UnconstrBackwardCorrection & robotoc::UnconstrBackwardCorrection::operator= ( const UnconstrBackwardCorrection & )

default

Default copy operator.

◆ operator=() [2/2]

UnconstrBackwardCorrection & robotoc::UnconstrBackwardCorrection::operator= ( UnconstrBackwardCorrection && )

defaultnoexcept

Default move assign operator.

◆ primalStepSize()

double robotoc::UnconstrBackwardCorrection::primalStepSize ( ) const

Gets the maximum primal step size of the fraction-to-boundary-rule.

Returns: The primal step size of the fraction-to-boundary-rule.

◆ setNumThreads()

void robotoc::UnconstrBackwardCorrection::setNumThreads ( const int nthreads )

Sets the number of threads of the parallel computations.

Parameters

[in] nthreads Number of the threads of the parallel computations. Must be positive.

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

/github/workspace/include/robotoc/parnmpc/unconstr_backward_correction.hpp

Public Member Functions

Detailed Description

Constructor & Destructor Documentation

◆ UnconstrBackwardCorrection() [1/4]

◆ UnconstrBackwardCorrection() [2/4]

◆ ~UnconstrBackwardCorrection()

◆ UnconstrBackwardCorrection() [3/4]

◆ UnconstrBackwardCorrection() [4/4]

Member Function Documentation

◆ backwardCorrection()

◆ coarseUpdate()

◆ dualStepSize()

◆ evalKKT()

◆ evalOCP()

◆ getEval()

◆ initAuxMat()

◆ initConstraints()

◆ integratePrimalSolution()

◆ integrateSolution()

◆ operator=() [1/2]

◆ operator=() [2/2]

◆ primalStepSize()

◆ setNumThreads()