gtsam::HybridNonlinearISAM Class Reference

Detailed Description

Wrapper class to manage ISAM in a nonlinear context.

Public Member Functions
Standard Constructors
	HybridNonlinearISAM (int reorderInterval=1, const HybridGaussianFactorGraph::Eliminate &eliminationFunction=HybridGaussianFactorGraph::EliminationTraitsType::DefaultEliminate)
	Periodically reorder and relinearize.
Standard Interface
Values	estimate ()
	Return the current solution estimate.
const HybridGaussianISAM &	bayesTree () const
	find the marginal covariance for a single variable
void	prune (const size_t maxNumberLeaves)
	Prune the underlying Bayes tree.
const Values &	getLinearizationPoint () const
	Return the current linearization point.
const DiscreteValues &	assignment () const
	Return the current discrete assignment.
const HybridNonlinearFactorGraph &	getFactorsUnsafe () const
	get underlying nonlinear graph
int	reorderInterval () const
	get counters
int	reorderCounter () const
	TODO: comment.
void	print (const std::string &s="", const KeyFormatter &keyFormatter=DefaultKeyFormatter) const
	prints out all contents of the system
void	printStats () const
	prints out clique statistics
void	saveGraph (const std::string &s, const KeyFormatter &keyFormatter=DefaultKeyFormatter) const
	saves the Tree to a text file in GraphViz format
Advanced Interface
void	update (const HybridNonlinearFactorGraph &newFactors, const Values &initialValues, const boost::optional< size_t > &maxNrLeaves=boost::none, const boost::optional< Ordering > &ordering=boost::none)
	Add new factors along with their initial linearization points.
void	reorder_relinearize ()
	Relinearization and reordering of variables.

Protected Attributes
gtsam::HybridGaussianISAM	isam_
	The internal iSAM object.
Values	linPoint_
	The current linearization point.
DiscreteValues	assignment_
	The discrete assignment.
HybridNonlinearFactorGraph	factors_
	The original factors, used when relinearizing.
int	reorderInterval_
	The reordering interval and counter.
int	reorderCounter_
HybridGaussianFactorGraph::Eliminate	eliminationFunction_
	The elimination function.

Constructor & Destructor Documentation

HybridNonlinearISAM()

gtsam::HybridNonlinearISAM::HybridNonlinearISAM ( int reorderInterval = 1, const HybridGaussianFactorGraph::Eliminate & eliminationFunction = HybridGaussianFactorGraph::EliminationTraitsType::DefaultEliminate )

inline

Periodically reorder and relinearize.

Parameters

reorderInterval

is the number of updates between reorderings, 0 never reorders (and is dangerous for memory consumption) 1 (default) reorders every time, in worse case is batch every update typical values are 50 or 100

Member Function Documentation

bayesTree()

const HybridGaussianISAM & gtsam::HybridNonlinearISAM::bayesTree ( ) const

inline

find the marginal covariance for a single variable

access the underlying bayes tree

prune()

void gtsam::HybridNonlinearISAM::prune ( const size_t maxNumberLeaves )

inline

Prune the underlying Bayes tree.

Parameters

maxNumberLeaves

The max number of leaf nodes to keep.

reorderInterval()

int gtsam::HybridNonlinearISAM::reorderInterval ( ) const

inline

get counters

TODO: comment

---

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

• /build/reproducible-path/gtsam-4.2.2+dfsg/gtsam/hybrid/HybridNonlinearISAM.h
• /build/reproducible-path/gtsam-4.2.2+dfsg/gtsam/hybrid/HybridNonlinearISAM.cpp