roboptim::trajectory::ProblemOverSplinesFactory< T, S > Class Template Reference

Factory of problems over splines. More...

#include <roboptim/trajectory/problem-over-splines-factory.hh>

List of all members.

Public Types
enum	CostType {   COST_DEFAULT,   COST_JERK }
	Cost functions supported. More...
typedef Problem< T >	problem_t
typedef problem_t::scaling_t	scaling_t
typedef problem_t::scalingVect_t	scalingVect_t
typedef problem_t::intervals_t	intervals_t
typedef problem_t::function_t	function_t
typedef problem_t::constraint_t	constraint_t
typedef std::vector< constraint_t >	constraints_t
typedef function_t::value_type	value_type
typedef function_t::size_type	size_type
typedef function_t::interval_t	interval_t
typedef function_t::vector_t	vector_t
typedef function_t::matrix_t	matrix_t
typedef ConstraintsOverSplines < T, S >	splinesConstraint_t
typedef boost::shared_ptr < splinesConstraint_t >	splinesConstraintPtr_t
typedef boost::tuple < splinesConstraintPtr_t, intervals_t, scaling_t >	globalConstraint_t
typedef GenericNumericLinearFunction < T >	numericLinearConstraint_t
typedef boost::shared_ptr < numericLinearConstraint_t >	numericLinearConstraintPtr_t
typedef boost::tuple < numericLinearConstraintPtr_t, interval_t, value_type >	freeze_t
typedef boost::variant < globalConstraint_t, freeze_t >	supportedConstraint_t
typedef S	spline_t
typedef boost::shared_ptr < spline_t >	splinePtr_t
typedef std::vector < boost::shared_ptr< S > >	splines_t
typedef std::vector< std::pair < value_type, std::vector < supportedConstraint_t > > >	problemConstraints_t
	The constraints are stored with their given startingPoint/time.
Public Member Functions
	ProblemOverSplinesFactory (const splines_t &splines, const problem_t &problem, CostType cost=COST_DEFAULT, value_type scaling=-1.)
	Constructor.
value_type	t0 () const
value_type &	t0 ()
value_type	tmax () const
value_type &	tmax ()
value_type	epsilon () const
value_type &	epsilon ()
void	updateRange (const interval_t &newRange, CostType cost=COST_DEFAULT)
	Updates the range of the optimization problem.
void	updateStartingPoint (value_type startingPoint, CostType cost=COST_DEFAULT)
	Updates the startingPoint of the problem, using updateRange.
void	updateEndingPoint (value_type endingPoint, CostType cost=COST_DEFAULT)
	Updates the endingPoint of the problem, using updateRange.
void	addSpline (const S &spline)
	Adds a spline to the problem.
void	addConstraint (value_type time, int order, const vector_t &value, const scaling_t &scaling, value_type eps=-1.)
	Adds freezing contraints on every spline at a given time.
void	addIntervalConstraint (value_type startingPoint, int order, const intervals_t &range, const scalingVect_t &scaling)
	Adds contraints on every spline starting at a given time.
void	addConstraint (value_type time, int order, const vector_t &value)
	Adds freezing contraints on every spline at a given time.
void	addIntervalConstraint (value_type startingPoint, int order, const intervals_t &range)
	Adds contraints on every spline starting at a given time.
const problem_t &	problem () const
	Return a reference to the stored problem.

---

Detailed Description

template<typename T, typename S>
class roboptim::trajectory::ProblemOverSplinesFactory< T, S >

Factory of problems over splines.

Template Parameters:

T	Matrix type.
S	Spline type.

---

Member Typedef Documentation

template<typename T, typename S>

typedef problem_t::constraint_t roboptim::trajectory::ProblemOverSplinesFactory< T, S >::constraint_t

template<typename T, typename S>

typedef std::vector<constraint_t> roboptim::trajectory::ProblemOverSplinesFactory< T, S >::constraints_t

template<typename T, typename S>

typedef boost::tuple<numericLinearConstraintPtr_t, interval_t, value_type> roboptim::trajectory::ProblemOverSplinesFactory< T, S >::freeze_t

template<typename T, typename S>

typedef problem_t::function_t roboptim::trajectory::ProblemOverSplinesFactory< T, S >::function_t

template<typename T, typename S>

typedef boost::tuple<splinesConstraintPtr_t, intervals_t, scaling_t> roboptim::trajectory::ProblemOverSplinesFactory< T, S >::globalConstraint_t

template<typename T, typename S>

typedef function_t::interval_t roboptim::trajectory::ProblemOverSplinesFactory< T, S >::interval_t

template<typename T, typename S>

typedef problem_t::intervals_t roboptim::trajectory::ProblemOverSplinesFactory< T, S >::intervals_t

template<typename T, typename S>

typedef function_t::matrix_t roboptim::trajectory::ProblemOverSplinesFactory< T, S >::matrix_t

template<typename T, typename S>

typedef GenericNumericLinearFunction<T> roboptim::trajectory::ProblemOverSplinesFactory< T, S >::numericLinearConstraint_t

template<typename T, typename S>

typedef boost::shared_ptr<numericLinearConstraint_t> roboptim::trajectory::ProblemOverSplinesFactory< T, S >::numericLinearConstraintPtr_t

template<typename T, typename S>

typedef Problem<T> roboptim::trajectory::ProblemOverSplinesFactory< T, S >::problem_t

template<typename T, typename S>

typedef std::vector<std::pair<value_type, std::vector<supportedConstraint_t> > > roboptim::trajectory::ProblemOverSplinesFactory< T, S >::problemConstraints_t

The constraints are stored with their given startingPoint/time.

template<typename T, typename S>

typedef problem_t::scaling_t roboptim::trajectory::ProblemOverSplinesFactory< T, S >::scaling_t

template<typename T, typename S>

typedef problem_t::scalingVect_t roboptim::trajectory::ProblemOverSplinesFactory< T, S >::scalingVect_t

template<typename T, typename S>

typedef function_t::size_type roboptim::trajectory::ProblemOverSplinesFactory< T, S >::size_type

template<typename T, typename S>

typedef S roboptim::trajectory::ProblemOverSplinesFactory< T, S >::spline_t

template<typename T, typename S>

typedef boost::shared_ptr<spline_t> roboptim::trajectory::ProblemOverSplinesFactory< T, S >::splinePtr_t

template<typename T, typename S>

typedef std::vector<boost::shared_ptr<S> > roboptim::trajectory::ProblemOverSplinesFactory< T, S >::splines_t

template<typename T, typename S>

typedef ConstraintsOverSplines<T, S> roboptim::trajectory::ProblemOverSplinesFactory< T, S >::splinesConstraint_t

template<typename T, typename S>

typedef boost::shared_ptr<splinesConstraint_t> roboptim::trajectory::ProblemOverSplinesFactory< T, S >::splinesConstraintPtr_t

template<typename T, typename S>

typedef boost::variant<globalConstraint_t, freeze_t> roboptim::trajectory::ProblemOverSplinesFactory< T, S >::supportedConstraint_t

template<typename T, typename S>

typedef function_t::value_type roboptim::trajectory::ProblemOverSplinesFactory< T, S >::value_type

template<typename T, typename S>

typedef function_t::vector_t roboptim::trajectory::ProblemOverSplinesFactory< T, S >::vector_t

---

Member Enumeration Documentation

template<typename T, typename S>

enum roboptim::trajectory::ProblemOverSplinesFactory::CostType

Cost functions supported.

Enumerator:

COST_DEFAULT
COST_JERK

---

Constructor & Destructor Documentation

template<typename T , typename S >

roboptim::trajectory::ProblemOverSplinesFactory< T, S >::ProblemOverSplinesFactory	(	const splines_t &	splines,
		const problem_t &	problem,
		CostType	cost = COST_DEFAULT,
		value_type	scaling = -1.
	)

Constructor.

Parameters:

splines	vector of splines considered.
problem	optimization problem.
cost	type of cost function.
scaling	scaling for the cost function (scaling > 0.).

---

Member Function Documentation

template<typename T , typename S >

void roboptim::trajectory::ProblemOverSplinesFactory< T, S >::addConstraint	(	value_type	time,
		int	order,
		const vector_t &	value,
		const scaling_t &	scaling,
		value_type	eps = -1.
	)

Adds freezing contraints on every spline at a given time.

These constraints are equality constraints on the value, derivative or second derivative (depending on the given order) of the spline at t=time. It constraints every spline.

Parameters:

time	Time of the constraint
order	Derivation order (0 for no derivation)
value	Goals of the constraints for each spline
scaling	Scalings of the constraints for each spline
eps	Epsilon used for the equality constraints

template<typename T , typename S >

void roboptim::trajectory::ProblemOverSplinesFactory< T, S >::addConstraint	(	value_type	time,
		int	order,
		const vector_t &	value
	)

Adds freezing contraints on every spline at a given time.

Calls the corresponding addConstraint with a scaling set to 1 for each spline.

template<typename T , typename S >

void roboptim::trajectory::ProblemOverSplinesFactory< T, S >::addIntervalConstraint	(	value_type	startingPoint,
		int	order,
		const intervals_t &	range,
		const scalingVect_t &	scaling
	)

Adds contraints on every spline starting at a given time.

These constraints are bounding constraints on the value, derivative or second derivative (depending on the given order) of the spline, on the time range between the given startingPoint and the end of the interval. It constraints every spline.

Parameters:

startingPoint	Time when the constraint interval starts
order	Derivation order (0 for no derivation)
range	Bounds of the constraint for each spline
scaling	Scalings of the constraints for each spline

template<typename T , typename S >

void roboptim::trajectory::ProblemOverSplinesFactory< T, S >::addIntervalConstraint	(	value_type	startingPoint,
		int	order,
		const intervals_t &	range
	)

Adds contraints on every spline starting at a given time.

Calls the corresponding addConstraint with a scaling set to 1 for each spline.

template<typename T , typename S>

void roboptim::trajectory::ProblemOverSplinesFactory< T, S >::addSpline

(

const S &

spline

)

Adds a spline to the problem.

Warning, the size of the existing cost function will not change, it can therefore break your optimization problem.

It is best practice to give all the splines to the constructor.

template<typename T , typename S >

ProblemOverSplinesFactory< T, S >::value_type roboptim::trajectory::ProblemOverSplinesFactory< T, S >::epsilon

(

)

const

template<typename T , typename S >

ProblemOverSplinesFactory< T, S >::value_type & roboptim::trajectory::ProblemOverSplinesFactory< T, S >::epsilon

(

)

template<typename T , typename S >

const ProblemOverSplinesFactory< T, S >::problem_t & roboptim::trajectory::ProblemOverSplinesFactory< T, S >::problem

(

)

const

Return a reference to the stored problem.

template<typename T , typename S >

ProblemOverSplinesFactory< T, S >::value_type roboptim::trajectory::ProblemOverSplinesFactory< T, S >::t0

(

)

const

template<typename T , typename S >

ProblemOverSplinesFactory< T, S >::value_type & roboptim::trajectory::ProblemOverSplinesFactory< T, S >::t0

(

)

template<typename T , typename S >

ProblemOverSplinesFactory< T, S >::value_type roboptim::trajectory::ProblemOverSplinesFactory< T, S >::tmax

(

)

const

template<typename T , typename S >

ProblemOverSplinesFactory< T, S >::value_type & roboptim::trajectory::ProblemOverSplinesFactory< T, S >::tmax

(

)

template<typename T , typename S >

void roboptim::trajectory::ProblemOverSplinesFactory< T, S >::updateEndingPoint	(	value_type	endingPoint,
		CostType	cost = COST_DEFAULT
	)

Updates the endingPoint of the problem, using updateRange.

template<typename T , typename S >

void roboptim::trajectory::ProblemOverSplinesFactory< T, S >::updateRange	(	const interval_t &	newRange,
		CostType	cost = COST_DEFAULT
	)

Updates the range of the optimization problem.

Parameters:

newRange	desired timeRange
cost	new function we want to use (or DEFAULT if keep the same one).

template<typename T , typename S >

void roboptim::trajectory::ProblemOverSplinesFactory< T, S >::updateStartingPoint	(	value_type	startingPoint,
		CostType	cost = COST_DEFAULT
	)

Updates the startingPoint of the problem, using updateRange.