Detailed Description

The ICP algorithm configuration data.

Definition at line 69 of file CICP.h.

#include <mrpt/slam/CICP.h>

Inheritance diagram for mrpt::slam::CICP::TConfigParams:

Public Member Functions
void	loadFromConfigFile (const mrpt::config::CConfigFileBase &source, const std::string &section) override
	This method load the options from a ".ini"-like file or memory-stored string list. More...

void	saveToConfigFile (mrpt::config::CConfigFileBase &c, const std::string &s) const override
	This method saves the options to a ".ini"-like file or memory-stored string list. More...

void	loadFromConfigFileName (const std::string &config_file, const std::string &section)
	Behaves like loadFromConfigFile, but you can pass directly a file name and a temporary CConfigFile object will be created automatically to load the file. More...

void	saveToConfigFileName (const std::string &config_file, const std::string &section) const
	Behaves like saveToConfigFile, but you can pass directly a file name and a temporary CConfigFile object will be created automatically to save the file. More...

void	dumpToConsole () const
	Just like dumpToTextStream() but sending the text to the console (std::cout) More...

virtual void	dumpToTextStream (std::ostream &out) const
	This method should clearly display all the contents of the structure in textual form, sending it to a std::ostream. More...

Public Attributes
double	thresholdDist {0.75}
	Initial threshold distance for two points to become a correspondence. More...

double	thresholdAng {0.15 * M_PI / 180.0}

double	ALFA {0.5}
	The scale factor for thresholds every time convergence is achieved. More...

double	smallestThresholdDist {0.1}
	The size for threshold such that iterations will stop, since it is considered precise enough. More...

double	covariance_varPoints {0.02 * 0.02}
	This is the normalization constant $\sigma^2_p$ that is used to scale the whole 3x3 covariance. More...

bool	doRANSAC {false}
	Perform a RANSAC step, mrpt::tfest::se2_l2_robust(), after the ICP convergence, to obtain a better estimation of the pose PDF. More...

double	kernel_rho {0.07}
	Cauchy kernel rho, for estimating the optimal transformation covariance, in meters (default = 0.07m) More...

bool	use_kernel {true}
	Whether to use kernel_rho to smooth distances, or use distances directly (default=true) More...

double	Axy_aprox_derivatives {0.05}
	[LM method only] The size of the perturbance in x & y used to estimate the Jacobians of the square error (default=0.05) More...

double	LM_initial_lambda {1e-4}
	[LM method only] The initial value of the lambda parameter in the LM method (default=1e-4) More...

bool	skip_cov_calculation {false}
	Skip the computation of the covariance (saves some time) (default=false) More...

bool	skip_quality_calculation {true}
	Skip the (sometimes) expensive evaluation of the term 'quality' at ICP output (Default=true) More...

uint32_t	corresponding_points_decimation {5}
	Decimation of the point cloud being registered against the reference one (default=5) - set to 1 to have the older (MRPT <0.9.5) behavior of not approximating ICP by ignoring the correspondence of some points. More...

Algorithms selection
TICPAlgorithm	ICP_algorithm {icpClassic}
	The algorithm to use (default: icpClassic). More...

TICPCovarianceMethod	ICP_covariance_method {icpCovFiniteDifferences}
	The method to use for covariance estimation (Default: icpCovFiniteDifferences) More...

Correspondence-finding criteria
bool	onlyUniqueRobust {false}

Termination criteria
if this option is enabled only the closest correspondence for each reference point will be kept (default=false).
unsigned int	maxIterations {40}
	Maximum number of iterations to run. More...

double	minAbsStep_trans {1e-6}
	If the correction in all translation coordinates (X,Y,Z) is below this threshold (in meters), iterations are terminated (Default:1e-6) More...

double	minAbsStep_rot {1e-6}
	If the correction in all rotation coordinates (yaw,pitch,roll) is below this threshold (in radians), iterations are terminated (Default:1e-6) More...

RANSAC-step options for mrpt::tfest::se2_l2_robust() if \a
doRANSAC=true
unsigned int	ransac_minSetSize {3}

unsigned int	ransac_maxSetSize {20}

unsigned int	ransac_nSimulations {100}

double	ransac_mahalanobisDistanceThreshold {3.0}

double	normalizationStd {0.02}
	RANSAC-step option: The standard deviation in X,Y of landmarks/points which are being matched (used to compute covariances in the SoG) More...

bool	ransac_fuseByCorrsMatch {true}

double	ransac_fuseMaxDiffXY {0.01}

double	ransac_fuseMaxDiffPhi {0.1 * M_PI / 180.0}

Static Protected Member Functions
static void	dumpVar_int (std::ostream &out, const char *varName, int v)
	Used to print variable info from dumpToTextStream with the macro LOADABLEOPTS_DUMP_VAR. More...

static void	dumpVar_float (std::ostream &out, const char *varName, float v)

static void	dumpVar_double (std::ostream &out, const char *varName, double v)

static void	dumpVar_bool (std::ostream &out, const char *varName, bool v)

static void	dumpVar_string (std::ostream &out, const char *varName, const std::string &v)

Member Function Documentation

◆ dumpToConsole()

void CLoadableOptions::dumpToConsole ( ) const

inherited

Just like dumpToTextStream() but sending the text to the console (std::cout)

Definition at line 43 of file CLoadableOptions.cpp.

References mrpt::config::CLoadableOptions::dumpToTextStream().

Referenced by mrpt::hmtslam::CTopLCDetector_GridMatching::computeTopologicalObservationModel(), mrpt::ros1bridge::MapHdl::loadMap(), mrpt::graphslam::deciders::CICPCriteriaNRD< GRAPH_T >::printParams(), and mrpt::apps::CGridMapAlignerApp::run().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ dumpToTextStream()

void CLoadableOptions::dumpToTextStream ( std::ostream & out ) const

virtualinherited

This method should clearly display all the contents of the structure in textual form, sending it to a std::ostream.

The default implementation in this base class relies on saveToConfigFile() to generate a plain text representation of all the parameters.

Definition at line 76 of file CLoadableOptions.cpp.

References mrpt::config::CConfigFileMemory::getContent(), out, and mrpt::config::CLoadableOptions::saveToConfigFile().

Referenced by mrpt::config::CLoadableOptions::dumpToConsole(), mrpt::maps::TMetricMapInitializer::dumpToTextStream(), mrpt::graphslam::deciders::CRangeScanOps< GRAPH_T >::TParams::dumpToTextStream(), mrpt::maps::COccupancyGridMap3D::TMapDefinition::dumpToTextStream_map_specific(), and mrpt::apps::ICP_SLAM_App_Base::run().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ dumpVar_bool()

void CLoadableOptions::dumpVar_bool	(	std::ostream &	out,
		const char *	varName,
		bool	v
	)

staticprotectedinherited

Definition at line 62 of file CLoadableOptions.cpp.

References mrpt::format(), LOADABLEOPTS_COLUMN_WIDTH, and out.

Here is the call graph for this function:

◆ dumpVar_double()

void CLoadableOptions::dumpVar_double	(	std::ostream &	out,
		const char *	varName,
		double	v
	)

staticprotectedinherited

Definition at line 56 of file CLoadableOptions.cpp.

References mrpt::format(), LOADABLEOPTS_COLUMN_WIDTH, and out.

Here is the call graph for this function:

◆ dumpVar_float()

void CLoadableOptions::dumpVar_float	(	std::ostream &	out,
		const char *	varName,
		float	v
	)

staticprotectedinherited

Definition at line 50 of file CLoadableOptions.cpp.

References mrpt::format(), LOADABLEOPTS_COLUMN_WIDTH, and out.

Here is the call graph for this function:

◆ dumpVar_int()

void CLoadableOptions::dumpVar_int	(	std::ostream &	out,
		const char *	varName,
		int	v
	)

staticprotectedinherited

Used to print variable info from dumpToTextStream with the macro LOADABLEOPTS_DUMP_VAR.

Definition at line 44 of file CLoadableOptions.cpp.

References mrpt::format(), LOADABLEOPTS_COLUMN_WIDTH, and out.

Here is the call graph for this function:

◆ dumpVar_string()

void CLoadableOptions::dumpVar_string	(	std::ostream &	out,
		const char *	varName,
		const std::string &	v
	)

staticprotectedinherited

Definition at line 69 of file CLoadableOptions.cpp.

References mrpt::format(), LOADABLEOPTS_COLUMN_WIDTH, and out.

Here is the call graph for this function:

◆ loadFromConfigFile()

void CICP::TConfigParams::loadFromConfigFile	(	const mrpt::config::CConfigFileBase &	source,
		const std::string &	section
	)

overridevirtual

This method load the options from a ".ini"-like file or memory-stored string list.

Only those parameters found in the given "section" and having the same name that the variable are loaded. Those not found in the file will stay with their previous values (usually the default values loaded at initialization). An example of an ".ini" file:

[section]
resolution    = 0.10   // blah blah...
modeSelection = 1      // 0=blah, 1=blah,...

See also: loadFromConfigFileName, saveToConfigFile

Implements mrpt::config::CLoadableOptions.

Definition at line 80 of file CICP.cpp.

References mrpt::DEG2RAD(), iniFile(), MRPT_LOAD_CONFIG_VAR, normalizationStd, mrpt::RAD2DEG(), and ransac_mahalanobisDistanceThreshold.

Referenced by mrpt::apps::ICP_SLAM_App_Base::run().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ loadFromConfigFileName()

void CLoadableOptions::loadFromConfigFileName	(	const std::string &	config_file,
		const std::string &	section
	)

inherited

Behaves like loadFromConfigFile, but you can pass directly a file name and a temporary CConfigFile object will be created automatically to load the file.

See also: loadFromConfigFile

Definition at line 22 of file CLoadableOptions.cpp.

References mrpt::config::CLoadableOptions::loadFromConfigFile().

Referenced by mrpt::graphslam::deciders::CICPCriteriaNRD< GRAPH_T >::loadParams().

Here is the call graph for this function:

Here is the caller graph for this function:

◆ saveToConfigFile()

void CICP::TConfigParams::saveToConfigFile	(	mrpt::config::CConfigFileBase &	target,
		const std::string &	section
	)		const

overridevirtual

This method saves the options to a ".ini"-like file or memory-stored string list.

See also: loadFromConfigFile, saveToConfigFileName

Reimplemented from mrpt::config::CLoadableOptions.

Definition at line 126 of file CICP.cpp.

References MRPT_SAVE_CONFIG_VAR_COMMENT, normalizationStd, and ransac_mahalanobisDistanceThreshold.

◆ saveToConfigFileName()

void CLoadableOptions::saveToConfigFileName	(	const std::string &	config_file,
		const std::string &	section
	)		const

inherited

Behaves like saveToConfigFile, but you can pass directly a file name and a temporary CConfigFile object will be created automatically to save the file.

See also: saveToConfigFile, loadFromConfigFileName

Definition at line 36 of file CLoadableOptions.cpp.

References mrpt::config::CLoadableOptions::saveToConfigFile().

Here is the call graph for this function:

Member Data Documentation

◆ ALFA

double mrpt::slam::CICP::TConfigParams::ALFA {0.5}

The scale factor for thresholds every time convergence is achieved.

Definition at line 117 of file CICP.h.

Referenced by ICPTests::align2scans(), and mrpt::hmtslam::CLSLAM_RBPF_2DLASER::prediction_and_update_pfOptimalProposal().

◆ Axy_aprox_derivatives

double mrpt::slam::CICP::TConfigParams::Axy_aprox_derivatives {0.05}

[LM method only] The size of the perturbance in x & y used to estimate the Jacobians of the square error (default=0.05)

Definition at line 158 of file CICP.h.

◆ corresponding_points_decimation

uint32_t mrpt::slam::CICP::TConfigParams::corresponding_points_decimation {5}

Decimation of the point cloud being registered against the reference one (default=5) - set to 1 to have the older (MRPT <0.9.5) behavior of not approximating ICP by ignoring the correspondence of some points.

The speed-up comes from a decimation of the number of KD-tree queries, the most expensive step in ICP

Definition at line 176 of file CICP.h.

◆ covariance_varPoints

double mrpt::slam::CICP::TConfigParams::covariance_varPoints {0.02 * 0.02}

This is the normalization constant $\sigma^2_p$ that is used to scale the whole 3x3 covariance.

This has a default value of $ (0.02)^2 $ , that is, a 2cm sigma. See paper: J.L. Blanco, J. Gonzalez-Jimenez, J.A. Fernandez-Madrigal, "A Robust, Multi-Hypothesis Approach to Matching Occupancy Grid Maps", Robotica, vol. 31, no. 5, pp. 687-701, 2013.

Definition at line 129 of file CICP.h.

◆ doRANSAC

bool mrpt::slam::CICP::TConfigParams::doRANSAC {false}

Perform a RANSAC step, mrpt::tfest::se2_l2_robust(), after the ICP convergence, to obtain a better estimation of the pose PDF.

Definition at line 133 of file CICP.h.

Referenced by ICPTests::align2scans(), and mrpt::hmtslam::CLSLAM_RBPF_2DLASER::prediction_and_update_pfOptimalProposal().

◆ ICP_algorithm

TICPAlgorithm mrpt::slam::CICP::TConfigParams::ICP_algorithm {icpClassic}

The algorithm to use (default: icpClassic).

See https://www.mrpt.org/tutorials/programming/scan-matching-and-icp/ for details

Definition at line 84 of file CICP.h.

Referenced by ICPTests::align2scans().

◆ ICP_covariance_method

TICPCovarianceMethod mrpt::slam::CICP::TConfigParams::ICP_covariance_method {icpCovFiniteDifferences}

The method to use for covariance estimation (Default: icpCovFiniteDifferences)

Definition at line 87 of file CICP.h.

◆ kernel_rho

double mrpt::slam::CICP::TConfigParams::kernel_rho {0.07}

Cauchy kernel rho, for estimating the optimal transformation covariance, in meters (default = 0.07m)

Definition at line 152 of file CICP.h.

◆ LM_initial_lambda

double mrpt::slam::CICP::TConfigParams::LM_initial_lambda {1e-4}

[LM method only] The initial value of the lambda parameter in the LM method (default=1e-4)

Definition at line 161 of file CICP.h.

◆ maxIterations

unsigned int mrpt::slam::CICP::TConfigParams::maxIterations {40}

Maximum number of iterations to run.

Definition at line 101 of file CICP.h.

Referenced by ICPTests::align2scans(), and mrpt::hmtslam::CLSLAM_RBPF_2DLASER::prediction_and_update_pfOptimalProposal().

◆ minAbsStep_rot

double mrpt::slam::CICP::TConfigParams::minAbsStep_rot {1e-6}

If the correction in all rotation coordinates (yaw,pitch,roll) is below this threshold (in radians), iterations are terminated (Default:1e-6)

Definition at line 109 of file CICP.h.

◆ minAbsStep_trans

double mrpt::slam::CICP::TConfigParams::minAbsStep_trans {1e-6}

If the correction in all translation coordinates (X,Y,Z) is below this threshold (in meters), iterations are terminated (Default:1e-6)

Definition at line 105 of file CICP.h.

◆ normalizationStd

double mrpt::slam::CICP::TConfigParams::normalizationStd {0.02}

RANSAC-step option: The standard deviation in X,Y of landmarks/points which are being matched (used to compute covariances in the SoG)

Definition at line 144 of file CICP.h.

◆ onlyUniqueRobust

bool mrpt::slam::CICP::TConfigParams::onlyUniqueRobust {false}

Definition at line 92 of file CICP.h.

◆ ransac_fuseByCorrsMatch

bool mrpt::slam::CICP::TConfigParams::ransac_fuseByCorrsMatch {true}

Definition at line 145 of file CICP.h.

◆ ransac_fuseMaxDiffPhi

double mrpt::slam::CICP::TConfigParams::ransac_fuseMaxDiffPhi {0.1 * M_PI / 180.0}

Definition at line 147 of file CICP.h.

◆ ransac_fuseMaxDiffXY

double mrpt::slam::CICP::TConfigParams::ransac_fuseMaxDiffXY {0.01}

Definition at line 146 of file CICP.h.

◆ ransac_mahalanobisDistanceThreshold

double mrpt::slam::CICP::TConfigParams::ransac_mahalanobisDistanceThreshold {3.0}

Definition at line 140 of file CICP.h.

◆ ransac_maxSetSize

unsigned int mrpt::slam::CICP::TConfigParams::ransac_maxSetSize {20}

Definition at line 138 of file CICP.h.

◆ ransac_minSetSize

unsigned int mrpt::slam::CICP::TConfigParams::ransac_minSetSize {3}

Definition at line 138 of file CICP.h.

◆ ransac_nSimulations

unsigned int mrpt::slam::CICP::TConfigParams::ransac_nSimulations {100}

Definition at line 139 of file CICP.h.

◆ skip_cov_calculation

bool mrpt::slam::CICP::TConfigParams::skip_cov_calculation {false}

Skip the computation of the covariance (saves some time) (default=false)

Definition at line 165 of file CICP.h.

◆ skip_quality_calculation

bool mrpt::slam::CICP::TConfigParams::skip_quality_calculation {true}

Skip the (sometimes) expensive evaluation of the term 'quality' at ICP output (Default=true)

Definition at line 168 of file CICP.h.

◆ smallestThresholdDist

double mrpt::slam::CICP::TConfigParams::smallestThresholdDist {0.1}

The size for threshold such that iterations will stop, since it is considered precise enough.

Definition at line 120 of file CICP.h.

Referenced by ICPTests::align2scans(), and mrpt::hmtslam::CLSLAM_RBPF_2DLASER::prediction_and_update_pfOptimalProposal().

◆ thresholdAng

double mrpt::slam::CICP::TConfigParams::thresholdAng {0.15 * M_PI / 180.0}

Definition at line 114 of file CICP.h.

Referenced by ICPTests::align2scans(), mrpt::hmtslam::CLSLAM_RBPF_2DLASER::prediction_and_update_pfOptimalProposal(), and TEST_F().

◆ thresholdDist

double mrpt::slam::CICP::TConfigParams::thresholdDist {0.75}

Initial threshold distance for two points to become a correspondence.

Definition at line 114 of file CICP.h.

Referenced by ICPTests::align2scans(), mrpt::hmtslam::CLSLAM_RBPF_2DLASER::prediction_and_update_pfOptimalProposal(), and TEST_F().

◆ use_kernel

bool mrpt::slam::CICP::TConfigParams::use_kernel {true}

Whether to use kernel_rho to smooth distances, or use distances directly (default=true)

Definition at line 155 of file CICP.h.

Detailed Description

Public Member Functions

Public Attributes

Static Protected Member Functions

Member Function Documentation

◆ dumpToConsole()

◆ dumpToTextStream()

◆ dumpVar_bool()

◆ dumpVar_double()

◆ dumpVar_float()

◆ dumpVar_int()

◆ dumpVar_string()

◆ loadFromConfigFile()

◆ loadFromConfigFileName()

◆ saveToConfigFile()

◆ saveToConfigFileName()

Member Data Documentation

◆ ALFA

◆ Axy_aprox_derivatives

◆ corresponding_points_decimation

◆ covariance_varPoints

◆ doRANSAC

◆ ICP_algorithm

◆ ICP_covariance_method

◆ kernel_rho

◆ LM_initial_lambda

◆ maxIterations

◆ minAbsStep_rot

◆ minAbsStep_trans

◆ normalizationStd

◆ onlyUniqueRobust

◆ ransac_fuseByCorrsMatch

◆ ransac_fuseMaxDiffPhi

◆ ransac_fuseMaxDiffXY

◆ ransac_mahalanobisDistanceThreshold

◆ ransac_maxSetSize

◆ ransac_minSetSize

◆ ransac_nSimulations

◆ skip_cov_calculation

◆ skip_quality_calculation

◆ smallestThresholdDist

◆ thresholdAng

◆ thresholdDist

◆ use_kernel