gtsam::GraduatedFactor Class Reference

Detailed Description

Graduated Factor for riSAM base class.

Inheritance diagram for gtsam::GraduatedFactor:

Public Member Functions
Public Interface
	GraduatedFactor (RobustLoss::shared_ptr loss, GraduationScheduler::shared_ptr scheduler)
	Constructor.
	GraduatedFactor (const GraduatedFactor &other)
	Copy constructor.
virtual GaussianFactor::shared_ptr	linearizeGraduated (const Values &currentEstimate) const =0
	Linearize this factor using the convexification parameter mu.
virtual double	residual (const Values &currentEstimate) const =0
	returns the residual of the factor
virtual double	robustLoss (const Values &currentEstimate) const =0
	returns the graduated robust loss \rho_\mu(r) of the factor
const RobustLoss::shared_ptr &	loss () const
	Returns the robust loss for this graduated factor.
const GraduationScheduler::shared_ptr &	scheduler () const
	Returns the graduation scheduler for this factor.
virtual NonlinearFactor::shared_ptr	cloneUngraduated () const =0
	Copies this factor as an instance of its base type without the graduated robust loss or scheduler.

Public Types
Types
typedef std::shared_ptr< GraduatedFactor >	shared_ptr
typedef noiseModel::mEstimator::Base	RobustLoss

Protected Attributes
Fields
RobustLoss::shared_ptr	robustLoss_
	The robust loss for this factor.
GraduationScheduler::shared_ptr	scheduler_
	The control param ($\mu$) scheduler for this factor.
std::shared_ptr< double >	mu_
	The unique mu control parameter for this factor.

Friends
class	RISAM
	Befriend RISAM to give access to these protected values.

Constructor & Destructor Documentation

GraduatedFactor()

gtsam::GraduatedFactor::GraduatedFactor ( RobustLoss::shared_ptr loss, GraduationScheduler::shared_ptr scheduler )

inline

Constructor.

Parameters

args	The arguments required to construct the FACTOR_TYPE
loss	The graduated robust loss applied to this factor
scheduler	The control param $\mu$ scheduler for this factor

Member Function Documentation

cloneUngraduated()

virtual NonlinearFactor::shared_ptr gtsam::GraduatedFactor::cloneUngraduated ( ) const

pure virtual

Copies this factor as an instance of its base type without the graduated robust loss or scheduler.

Implemented in gtsam::GenericGraduatedFactor< FACTOR_TYPE >.

linearizeGraduated()

virtual GaussianFactor::shared_ptr gtsam::GraduatedFactor::linearizeGraduated ( const Values & currentEstimate ) const

pure virtual

Linearize this factor using the convexification parameter mu.

Parameters

currentEstimate

the estimate at which to linearize the factor

Implemented in gtsam::GenericGraduatedFactor< FACTOR_TYPE >.

residual()

virtual double gtsam::GraduatedFactor::residual ( const Values & currentEstimate ) const

pure virtual

returns the residual of the factor

Implemented in gtsam::GenericGraduatedFactor< FACTOR_TYPE >.

robustLoss()

virtual double gtsam::GraduatedFactor::robustLoss ( const Values & currentEstimate ) const

pure virtual

returns the graduated robust loss \rho_\mu(r) of the factor

Implemented in gtsam::GenericGraduatedFactor< FACTOR_TYPE >.

---

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

• /tmp/gtsam-4.3.0-doxygen.rsXPUS/source/gtsam/sam/RISAMGraduatedFactor.h