OptimizationAlgorithmDogleg< t_solver >

Implementation of Powell's Dogleg Algorithm. More...

Inheritance diagram for OptimizationAlgorithmDogleg< t_solver >:

Collaboration diagram for OptimizationAlgorithmDogleg< t_solver >:

Public Types
enum
	type of the step to take
Public Types inherited from OptimizationAlgorithm
enum	SolverResult { Terminate =2 , OK =1 , Fail =-1 }
	Result codes returned by the solver after each iteration. More...

Public Member Functions
	OptimizationAlgorithmDogleg ()
	construct the Dogleg algorithm, which will use the given Solver for solving the linearized system.
int	lastStep () const
	return the type of the last step taken by the algorithm
virtual void	printVerbose (std::ostream &os) const
	Prints verbose information about the current Dogleg state.
virtual OptimizationAlgorithm::SolverResult	solve (int iteration, bool online, int *blockPoseIndices, int *blockLandmarkIndices)
	Performs a single Dogleg iteration.
g_type	trustRegion () const
	return the diameter of the trust region
Public Member Functions inherited from OptimizationAlgorithmWithHessian< t_solver >
	OptimizationAlgorithmWithHessian ()
	Default constructor.
void	addEdge (HyperGraph::HGEdge *e)
	Adds an edge to the optimizer.
void	addVertex (HyperGraph::HGVertex &v)
	Adds a vertex to the optimizer.
bool	buildLinearStructure (IndexScratch &scratch)
	Builds the linear system structure.
void	clear ()
	Clears the solver state.
bool	init (bool online=false)
	Initializes the solver, detecting whether Schur complement should be used.
SparseOptimizer &	optimizer ()
	Returns a mutable reference to the underlying sparse optimizer.
const SparseOptimizer &	optimizer () const
	Returns a const reference to the underlying sparse optimizer.
void	updateLinearSystem ()
	Rebuilds the linear system (Hessian and gradient) from the current graph state.
bool	updateStructure (const Buffer< HyperGraph::HGVertex * > &vset, const Set< HyperGraph::HGVertex * > &edges)
	Updates the linear system structure after graph topology changes.

Static Public Member Functions
static const char *	stepType2Str (int stepType)
	convert the type into an integer

Protected Attributes
Eigen::VectorX< g_type >	_auxVector
	auxilary vector used to perform multiplications or other stuff
g_type	_currentLambda
	The damping factor to force positive definite matrix.
g_type	_delta
	Trust region radius.
Eigen::VectorX< g_type >	_hdl
	final dogleg step
Eigen::VectorX< g_type >	_hsd
	steepest decent step
g_type	_initialLambda
	Initial damping factor for positive-definite enforcement.
g_type	_lamdbaFactor
	Multiplicative factor for adjusting lambda.
int	_lastNumTries
	Number of tries in the last iteration.
int	_lastStep
	Type of the step taken by the algorithm.
int	_maxTrialsAfterFailure
	Maximum number of trials before declaring failure.
g_type	_userDeltaInit
	User-specified initial trust region radius.
bool	_wasPDInAllIterations
	Whether the matrix was positive definite in all iterations.

Additional Inherited Members
Public Attributes inherited from OptimizationAlgorithmWithHessian< t_solver >
t_solver	solver
	The underlying solver instance.

Detailed Description

template<class t_solver>
class OptimizationAlgorithmDogleg< t_solver >

Implementation of Powell's Dogleg Algorithm.

Definition at line 41 of file optimization_algorithm_dogleg.h.

Member Function Documentation

printVerbose()

template<class t_solver>

virtual void OptimizationAlgorithmDogleg< t_solver >::printVerbose ( std::ostream & os ) const

inlinevirtual

Prints verbose information about the current Dogleg state.

Parameters

[in]

os

The output stream.

Definition at line 217 of file optimization_algorithm_dogleg.h.

References _currentLambda, _delta, _lastNumTries, _lastStep, _wasPDInAllIterations, and stepType2Str().

solve()

template<class t_solver>

virtual OptimizationAlgorithm::SolverResult OptimizationAlgorithmDogleg< t_solver >::solve ( int iteration, bool online, int * blockPoseIndices, int * blockLandmarkIndices )

inlinevirtual

Performs a single Dogleg iteration.

Parameters

[in]	iteration	Current iteration index.
[in]	online	Whether this is an online step.
[in]	blockPoseIndices	Pose block index array.
[in]	blockLandmarkIndices	Landmark block index array.

Returns

The solver result (OK, Fail, or Terminate).

Definition at line 73 of file optimization_algorithm_dogleg.h.

References OptimizationAlgorithmWithHessian< t_solver >::OptimizationAlgorithmWithHessian(), _auxVector, _currentLambda, _delta, _hdl, _hsd, _initialLambda, _lamdbaFactor, _lastNumTries, _lastStep, _maxTrialsAfterFailure, _userDeltaInit, _wasPDInAllIterations, OptimizationAlgorithm::Fail, getMax(), OptimizationAlgorithm::OK, OptimizationAlgorithmWithHessian< t_solver >::optimizer(), OptimizationAlgorithmWithHessian< t_solver >::solver, sqrt(), and OptimizationAlgorithm::Terminate.

---

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

• GraphOptimization/Headers/optimization_algorithm_dogleg.h