roboptim::trajectory::ConstrainedBSpline< N > Class Template Reference

Constrained B-spline. More...

#include <roboptim/trajectory/constrained-b-spline.hh>

Inheritance diagram for roboptim::trajectory::ConstrainedBSpline< N >:

List of all members.

Public Types
typedef parent_t::interval_t	interval_t
	Import interval type.
Public Member Functions
	ROBOPTIM_NTIMES_DERIVABLE_FUNCTION_FWD_TYPEDEFS_ (BSpline< N >)
	Parent type and imports.
	ConstrainedBSpline (interval_t timeRange, size_type dimension, const vector_t &parameters, const std::string name="Constrained B-Spline")
	see B-Spline constructors for documentation
	ConstrainedBSpline (interval_t tr, size_type dimension, const vector_t &parameters, const_vector_ref knots, std::string name="Constrained B-Spline")
	see B-Spline constructors for documentation
virtual	~ConstrainedBSpline ()
	Destructor of constrained B-spline.
void	addFixedConstraint (double t, size_type dimension, value_type value, size_type derivative=0)
	Creates a constraint on the basic spline.
void	addCoupledConstraint (value_type t_1, size_type dimension_1, value_type t_2, size_type dimension_2, size_type derivative=0, value_type factor=1.)
	Create a constraint against another part of the spline.
const vector_t &	parameters () const
	Overloaded parameters method from Trajectory<N>.
void	setParameters (const_vector_ref)
	Overloaded setParameters method from Trajectory<N>.
virtual Trajectory< N > *	resize (interval_t timeRange) const
	Clone and resize a trajectory.
jacobian_t	variationDerivWrtParam (double t, size_type order) const
	Get the variation of a derivative with respect to parameter vector.
Protected Member Functions
void	updateProjector ()
	Update the projector matrix. Called after adding a constraint.
Protected Attributes
matrix_t	constraints_
	Matrix storing constraints.
vector_t	constraint_values_
	Constraint values.
vector_t	tunables_
	Tunable parameters, i.e. the free parameters.
matrix_t	projector_
	Projector matrix.
vector_t	projector_offset_
	Result of: Constraints * X = Constraint values.

---

Detailed Description

template<int N>
class roboptim::trajectory::ConstrainedBSpline< N >

Constrained B-spline.

Template Parameters:

N	B-spline order.

---

Member Typedef Documentation

template<int N>

typedef parent_t::interval_t roboptim::trajectory::ConstrainedBSpline< N >::interval_t

Import interval type.

Reimplemented from roboptim::trajectory::BSpline< N >.

---

Constructor & Destructor Documentation

template<int N>

roboptim::trajectory::ConstrainedBSpline< N >::ConstrainedBSpline	(	interval_t	timeRange,
		size_type	dimension,
		const vector_t &	parameters,
		const std::string	name = "Constrained B-Spline"
	)

see B-Spline constructors for documentation

template<int N>

roboptim::trajectory::ConstrainedBSpline< N >::ConstrainedBSpline	(	interval_t	tr,
		size_type	dimension,
		const vector_t &	parameters,
		const_vector_ref	knots,
		std::string	name = "Constrained B-Spline"
	)

see B-Spline constructors for documentation

template<int N>

roboptim::trajectory::ConstrainedBSpline< N >::~ConstrainedBSpline

(

)

[virtual]

Destructor of constrained B-spline.

---

Member Function Documentation

template<int N>

void roboptim::trajectory::ConstrainedBSpline< N >::addCoupledConstraint	(	value_type	t_1,
		size_type	dimension_1,
		value_type	t_2,
		size_type	dimension_2,
		size_type	derivative = 0,
		value_type	factor = 1.
	)

Create a constraint against another part of the spline.

\(\frac{\partial^d}{\partial t^d} f_{d_1}(t_1) = factor * \frac{\partial^d}{\partial t^d} f_{d_2}(t_2) \)

Parameters:

t_1	first time.
dimension_1	first dimension.
t_2	second time.
dimension_2	second dimension.
derivative	derivative order ( \(d\)).
factor	scalar factor.

References roboptim::trajectory::detail::fixTime().

template<int N>

void roboptim::trajectory::ConstrainedBSpline< N >::addFixedConstraint	(	double	t,
		size_type	dimension,
		value_type	value,
		size_type	derivative = 0
	)

Creates a constraint on the basic spline.

This reduces the number of parameter by one. The constraint equation is: \(value = \frac{\partial^derivative}{\partial t^derivative} f_{dimension}(t) \)

Parameters:

t	time in the spline to constrain.
dimension	Which dimension of the spline to constrain.
value	Desired spline value at t.
derivative	Which derivative of the spline to constrain.

References roboptim::trajectory::detail::fixTime().

template<int N>

const ConstrainedBSpline< N >::vector_t & roboptim::trajectory::ConstrainedBSpline< N >::parameters

(

)

const

Overloaded parameters method from Trajectory<N>.

Returns only the tunable parameters.

Reimplemented from roboptim::trajectory::Trajectory< N >.

template<int N>

Trajectory< N > * roboptim::trajectory::ConstrainedBSpline< N >::resize

(

interval_t

timeRange

)

const [virtual]

Clone and resize a trajectory.

Parameters:

timeRange

new time range.

Returns:

cloned trajectory with new time range.

Reimplemented from roboptim::trajectory::BSpline< N >.

References roboptim::trajectory::ConstrainedBSpline< N >::constraint_values_, roboptim::trajectory::ConstrainedBSpline< N >::constraints_, and roboptim::trajectory::ConstrainedBSpline< N >::tunables_.

template<int N>

roboptim::trajectory::ConstrainedBSpline< N >::ROBOPTIM_NTIMES_DERIVABLE_FUNCTION_FWD_TYPEDEFS_

(

BSpline< N >

)

Parent type and imports.

template<int N>

void roboptim::trajectory::ConstrainedBSpline< N >::setParameters

(

const_vector_ref

parameters_in

)

[virtual]

Overloaded setParameters method from Trajectory<N>.

Given the free parameters, it calculates the spline parameters.

Reimplemented from roboptim::trajectory::BSpline< N >.

template<int N>

void roboptim::trajectory::ConstrainedBSpline< N >::updateProjector

(

)

[protected]

Update the projector matrix. Called after adding a constraint.

template<int N>

ConstrainedBSpline< N >::jacobian_t roboptim::trajectory::ConstrainedBSpline< N >::variationDerivWrtParam	(	double	t,
		size_type	order
	)		const [virtual]

Get the variation of a derivative with respect to parameter vector.

Parameters:

t	value \(t\) in the definition interval.
order	order \(r\) of the derivative.

Returns:

jacobian

\[ \frac{\partial}{\partial\textbf{p}} \left(\frac{d^r\Gamma_{\textbf{p}}}{dt^r}(t)\right) \]

Implements roboptim::trajectory::Trajectory< N >.

References roboptim::trajectory::Polynomial< N >::derivative().

---

Member Data Documentation

template<int N>

vector_t roboptim::trajectory::ConstrainedBSpline< N >::constraint_values_ [protected]

Constraint values.

Referenced by roboptim::trajectory::ConstrainedBSpline< N >::resize().

template<int N>

matrix_t roboptim::trajectory::ConstrainedBSpline< N >::constraints_ [protected]

Matrix storing constraints.

One row per constraint, one column per parameter.

Referenced by roboptim::trajectory::ConstrainedBSpline< N >::resize().

template<int N>

matrix_t roboptim::trajectory::ConstrainedBSpline< N >::projector_ [protected]

Projector matrix.

template<int N>

vector_t roboptim::trajectory::ConstrainedBSpline< N >::projector_offset_ [protected]

Result of: Constraints * X = Constraint values.

template<int N>

vector_t roboptim::trajectory::ConstrainedBSpline< N >::tunables_ [protected]

Tunable parameters, i.e. the free parameters.

Referenced by roboptim::trajectory::ConstrainedBSpline< N >::resize().