roboptim::trajectory::StateFunction< T > Class Template Reference

Trajectory cost function defined by state evaluation at parameter. More...

#include <roboptim/trajectory/state-function.hh>

Inheritance diagram for roboptim::trajectory::StateFunction< T >:

Public Types
typedef T	trajectory_t
	Trajectory type. More...

Public Member Functions
	StateFunction (const trajectory_t &gamma, boost::shared_ptr< DerivableFunction > cost, const StableTimePoint tpt, size_type order=1)
	Constructor. More...
virtual	~StateFunction ()
size_type	order () const

Static Public Member Functions
template<typename M >
static void	addToProblem (const trajectory_t &trajectory, boost::shared_ptr< DerivableFunction > function, unsigned order, Problem< M > &problem, typename Function::interval_t bounds, unsigned nConstraints)
template<typename M >
static void	addToProblem (const trajectory_t &trajectory, boost::shared_ptr< DerivableFunction > function, unsigned order, Problem< M > &problem, const typename Function::intervals_t &bounds, const std::vector< typename Function::value_type > &scales, unsigned nConstraints)

Protected Member Functions
void	impl_compute (result_ref, const_argument_ref) const
void	impl_gradient (gradient_ref, const_argument_ref, size_type) const

Detailed Description

template<typename T>
class roboptim::trajectory::StateFunction< T >

Trajectory cost function defined by state evaluation at parameter.

The state along a trajectory is defined as the vector containing the configuration and derivatives up to order \(r\) of the configuration.

\[ \textbf{Cost}(\Gamma) = cost \left({\Gamma(t)}, {\dot{\Gamma}(t)},\cdots,\frac{d^{r}\Gamma}{dt^{r}}(t) \right) \]

where

• \(\textbf{Cost}\) is the trajectory cost,
• \(cost\) is the state cost,
• \(t\) is the parameter along the trajectory where the cost is evaluated (fixed at construction),
• \(r\) is called the order of the state.

Template Parameters

T	trajectory type

Member Typedef Documentation

template<typename T >

typedef T roboptim::trajectory::StateFunction< T >::trajectory_t

Trajectory type.

Constructor & Destructor Documentation

template<typename T >

roboptim::trajectory::StateFunction< T >::StateFunction	(	const trajectory_t &	gamma,
		boost::shared_ptr< DerivableFunction >	cost,
		const StableTimePoint	tpt,
		size_type	order = 1
	)

Constructor.

Parameters

gamma	Trajectory \(\Gamma\) along which the state is evaluated.
cost	state cost: \(cost\).
tpt	parameter \(t\) where the state is evaluated.
order	order \(r\) of derivation.

Exceptions

std::runtime_error

template<typename T >

roboptim::trajectory::StateFunction< T >::~StateFunction ( )

virtual

Member Function Documentation

template<typename T >

template<typename M >

static void roboptim::trajectory::StateFunction< T >::addToProblem ( const trajectory_t &  trajectory, boost::shared_ptr< DerivableFunction >  function, unsigned  order, Problem< M > &  problem, typename Function::interval_t  bounds, unsigned  nConstraints  )

inlinestatic

Template Parameters

M	Eigen matrix type

References roboptim::trajectory::tMax.

template<typename T >

template<typename M >

static void roboptim::trajectory::StateFunction< T >::addToProblem ( const trajectory_t &  trajectory, boost::shared_ptr< DerivableFunction >  function, unsigned  order, Problem< M > &  problem, const typename Function::intervals_t &  bounds, const std::vector< typename Function::value_type > &  scales, unsigned  nConstraints  )

inlinestatic

References roboptim::trajectory::tMax.

template<typename T >

void roboptim::trajectory::StateFunction< T >::impl_compute ( result_ref  res, const_argument_ref  p  ) const

protected

ROBOPTIM_DO_NOT_CHECK_ALLOCATION

ROBOPTIM_DO_NOT_CHECK_ALLOCATION

template<typename T >

void roboptim::trajectory::StateFunction< T >::impl_gradient ( gradient_ref  grad, const_argument_ref  p, size_type  i  ) const

protected

ROBOPTIM_DO_NOT_CHECK_ALLOCATION

ROBOPTIM_DO_NOT_CHECK_ALLOCATION

template<typename T >

StateFunction< T >::size_type roboptim::trajectory::StateFunction< T >::order

(

)

const