g2o::OptimizationAlgorithm Class Reference

Generic interface for a non-linear solver operating on a graph. More...

#include <optimization_algorithm.h>

Inheritance diagram for g2o::OptimizationAlgorithm:

Collaboration diagram for g2o::OptimizationAlgorithm:

Public Member Functions
	OptimizationAlgorithm ()
virtual	~OptimizationAlgorithm ()
virtual bool	init (bool online=false)=0
virtual SolverResult	solve (int iteration, bool online=false)=0
virtual bool	computeMarginals (SparseBlockMatrix< MatrixX > &spinv, const std::vector< std::pair< int, int > > &blockIndices)=0
virtual bool	updateStructure (const std::vector< HyperGraph::Vertex * > &vset, const HyperGraph::EdgeSet &edges)=0
virtual void	printVerbose (std::ostream &os) const
const SparseOptimizer *	optimizer () const
	return the optimizer operating on
SparseOptimizer *	optimizer ()
void	setOptimizer (SparseOptimizer *optimizer)
const PropertyMap &	properties () const
	return the properties of the solver
bool	updatePropertiesFromString (const std::string &propString)
void	printProperties (std::ostream &os) const

Protected Attributes
SparseOptimizer *	_optimizer
	the optimizer the solver is working on
PropertyMap	_properties

Private Member Functions
	OptimizationAlgorithm (const OptimizationAlgorithm &)
OptimizationAlgorithm &	operator= (const OptimizationAlgorithm &)

Detailed Description

Generic interface for a non-linear solver operating on a graph.

Definition at line 46 of file optimization_algorithm.h.

Constructor & Destructor Documentation

OptimizationAlgorithm() [1/2]

g2o::OptimizationAlgorithm::OptimizationAlgorithm

(

)

Definition at line 33 of file optimization_algorithm.cpp.

: _optimizer(nullptr) {}

~OptimizationAlgorithm()

g2o::OptimizationAlgorithm::~OptimizationAlgorithm ( )

virtual

Definition at line 35 of file optimization_algorithm.cpp.

{}

OptimizationAlgorithm() [2/2]

g2o::OptimizationAlgorithm::OptimizationAlgorithm ( const OptimizationAlgorithm &  )

inlineprivate

Definition at line 115 of file optimization_algorithm.h.

{}

Member Function Documentation

computeMarginals()

virtual bool g2o::OptimizationAlgorithm::computeMarginals ( SparseBlockMatrix< MatrixX > &  spinv, const std::vector< std::pair< int, int > > &  blockIndices  )

pure virtual

computes the block diagonal elements of the pattern specified in the input and stores them in given SparseBlockMatrix. If your solver does not support computing the marginals, return false.

Implemented in g2o::StructureOnlySolver< PointDoF >, and g2o::OptimizationAlgorithmWithHessian.

Referenced by g2o::SparseOptimizer::computeMarginals().

init()

virtual bool g2o::OptimizationAlgorithm::init ( bool  online = false )

pure virtual

initialize the solver, called once before the first call to solve()

Implemented in g2o::OptimizationAlgorithmWithHessian, and g2o::StructureOnlySolver< PointDoF >.

Referenced by g2o::computeSimpleStars(), g2o::Star::labelStarEdges(), g2o::SparseOptimizer::optimize(), g2o::SparseOptimizerIncremental::optimize(), and g2o::SparseOptimizerOnline::optimize().

operator=()

OptimizationAlgorithm & g2o::OptimizationAlgorithm::operator= ( const OptimizationAlgorithm &  )

inlineprivate

Definition at line 116 of file optimization_algorithm.h.

                                                                 {
    return *this;
  }

optimizer() [1/2]

SparseOptimizer * g2o::OptimizationAlgorithm::optimizer ( )

inline

Definition at line 88 of file optimization_algorithm.h.

{ return _optimizer; }

optimizer() [2/2]

const SparseOptimizer * g2o::OptimizationAlgorithm::optimizer ( ) const

inline

return the optimizer operating on

Definition at line 87 of file optimization_algorithm.h.

{ return _optimizer; }

Referenced by g2o::StructureOnlySolver< PointDoF >::init(), and setOptimizer().

printProperties()

void g2o::OptimizationAlgorithm::printProperties

(

std::ostream &

os

)

const

print the properties to a stream in a human readable fashion

Definition at line 37 of file optimization_algorithm.cpp.

                                                                {
  os << "------------- Algorithm Properties -------------" << endl;
  for (PropertyMap::const_iterator it = _properties.begin();
       it != _properties.end(); ++it) {
    BaseProperty* p = it->second;
    os << it->first << "\t" << p->toString() << endl;
  }
  os << "------------------------------------------------" << endl;
}

References _properties, and g2o::BaseProperty::toString().

Referenced by main().

printVerbose()

virtual void g2o::OptimizationAlgorithm::printVerbose ( std::ostream &  os ) const

inlinevirtual

called by the optimizer if verbose. re-implement, if you want to print something

Reimplemented in g2o::OptimizationAlgorithmDogleg, g2o::OptimizationAlgorithmGaussNewton, and g2o::OptimizationAlgorithmLevenberg.

Definition at line 83 of file optimization_algorithm.h.

{ (void)os; };

Referenced by g2o::SparseOptimizer::optimize().

properties()

const PropertyMap & g2o::OptimizationAlgorithm::properties ( ) const

inline

return the properties of the solver

Definition at line 96 of file optimization_algorithm.h.

{ return _properties; }

Referenced by MainWindow::on_btnOptimizerParameters_clicked().

setOptimizer()

void g2o::OptimizationAlgorithm::setOptimizer

(

SparseOptimizer *

optimizer

)

specify on which optimizer the solver should work on

Definition at line 52 of file optimization_algorithm.cpp.

                                                                   {
  _optimizer = optimizer;
}

References _optimizer, and optimizer().

Referenced by g2o::SparseOptimizer::setAlgorithm().

solve()

virtual SolverResult g2o::OptimizationAlgorithm::solve ( int  iteration, bool  online = false  )

pure virtual

Solve one iteration. The SparseOptimizer running on-top will call this for the given number of iterations.

Parameters

iteration

indicates the current iteration

Implemented in g2o::OptimizationAlgorithmDogleg, g2o::OptimizationAlgorithmGaussNewton, g2o::OptimizationAlgorithmLevenberg, g2o::SolverSLAM2DLinear, and g2o::StructureOnlySolver< PointDoF >.

Referenced by g2o::SparseOptimizer::optimize().

updatePropertiesFromString()

bool g2o::OptimizationAlgorithm::updatePropertiesFromString

(

const std::string &

propString

)

update the properties from a string, see PropertyMap::updateMapFromString()

Definition at line 47 of file optimization_algorithm.cpp.

                                 {
  return _properties.updateMapFromString(propString);
}

References _properties, and g2o::PropertyMap::updateMapFromString().

Referenced by main().

updateStructure()

virtual bool g2o::OptimizationAlgorithm::updateStructure ( const std::vector< HyperGraph::Vertex * > &  vset, const HyperGraph::EdgeSet &  edges  )

pure virtual

update the structures for online processing

Implemented in g2o::StructureOnlySolver< PointDoF >, and g2o::OptimizationAlgorithmWithHessian.

Referenced by g2o::SparseOptimizer::updateInitialization(), and g2o::SparseOptimizerIncremental::updateInitialization().

Member Data Documentation

_optimizer

SparseOptimizer* g2o::OptimizationAlgorithm::_optimizer

protected

the optimizer the solver is working on

Definition at line 109 of file optimization_algorithm.h.

Referenced by g2o::OptimizationAlgorithmLevenberg::computeLambdaInit(), g2o::OptimizationAlgorithmWithHessian::init(), setOptimizer(), g2o::OptimizationAlgorithmDogleg::solve(), g2o::OptimizationAlgorithmGaussNewton::solve(), g2o::OptimizationAlgorithmLevenberg::solve(), and g2o::SolverSLAM2DLinear::solveOrientation().

_properties

PropertyMap g2o::OptimizationAlgorithm::_properties

protected

the properties of your solver, use this to store the parameters of your solver

Definition at line 110 of file optimization_algorithm.h.

Referenced by g2o::OptimizationAlgorithmDogleg::OptimizationAlgorithmDogleg(), g2o::OptimizationAlgorithmLevenberg::OptimizationAlgorithmLevenberg(), g2o::OptimizationAlgorithmWithHessian::OptimizationAlgorithmWithHessian(), printProperties(), and updatePropertiesFromString().

---

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

• /build/reproducible-path/g2o-0~20230806/g2o/core/optimization_algorithm.h
• /build/reproducible-path/g2o-0~20230806/g2o/core/optimization_algorithm.cpp