Detailed Description

The ICP algorithm configuration data.

#include <mrpt/slam/CICP.h>

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

Public Member Functions
	TConfigParams ()
	Initializer for default values:
void	loadFromConfigFile (const mrpt::utils::CConfigFileBase &source, const std::string &section)
	See utils::CLoadableOptions.
void	dumpToTextStream (CStream &out) const
	See utils::CLoadableOptions.
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.
virtual void	saveToConfigFile (mrpt::utils::CConfigFileBase &source, const std::string &section)
	This method saves the options to a ".ini"-like file or memory-stored string list.
void	saveToConfigFileName (const std::string &config_file, const std::string &section)
	Behaves like saveToConfigFile, but you can pass directly a file name and a temporary CConfigFile object will be created automatically to save the file.
void	dumpToConsole () const
	This method must display clearly all the contents of the structure in textual form, sending it to a CStream.
Public Attributes
TICPAlgorithm	ICP_algorithm
	The algorithm to use (default: icpClassic) See http://www.mrpt.org/Scan_Matching_Algorithms for details.
bool	onlyClosestCorrespondences
	The usual approach: to consider only the closest correspondence for each local point (Default to true)
bool	onlyUniqueRobust
float	thresholdDist
float	thresholdAng
	Initial threshold distance for two points to become a correspondence.
float	ALFA
	The scale factor for threshold everytime convergence is achieved.
float	smallestThresholdDist
	The size for threshold such that iterations will stop, since it is considered precise enough.
float	covariance_varPoints
	This is the normalization constant $\sigma^2_p$ that is used to scale the whole 3x3 covariance.
bool	doRANSAC
	Perform a RANSAC step after the ICP convergence, to obtain a better estimation of the pose PDF.
unsigned int	ransac_minSetSize
	RANSAC-step options:
unsigned int	ransac_maxSetSize
unsigned int	ransac_nSimulations
float	ransac_mahalanobisDistanceThreshold
	RANSAC-step options:
float	normalizationStd
	RANSAC-step option: The standard deviation in X,Y of landmarks/points which are being matched (used to compute covariances in the SoG)
bool	ransac_fuseByCorrsMatch
	RANSAC-step options:
float	ransac_fuseMaxDiffXY
	RANSAC-step options:
float	ransac_fuseMaxDiffPhi
float	kernel_rho
	Cauchy kernel rho, for estimating the optimal transformation covariance, in meters (default = 0.07m).
bool	use_kernel
	Whether to use kernel_rho to smooth distances, or use distances directly (default=true)
float	Axy_aprox_derivatives
	The size of the perturbance in x & y used to estimate the Jacobians of the square error (in LM & IKF methods, default=0.05).
float	LM_initial_lambda
	The initial value of the lambda parameter in the LM method (default=1e-4).
bool	skip_cov_calculation
	Skip the computation of the covariance (saves some time) (default=false)
uint32_t	corresponding_points_decimation
	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.
Termination criteria
Apart of "onlyClosestCorrespondences=true", if this option is enabled only the closest correspondence for each reference point will be kept (default=false).
unsigned int	maxIterations
	Maximum number of iterations to run.
float	minAbsStep_trans
	If the correction in all translation coordinates (X,Y,Z) is below this threshold (in meters), iterations are terminated (Default:1e-6)
float	minAbsStep_rot
	If the correction in all rotation coordinates (yaw,pitch,roll) is below this threshold (in radians), iterations are terminated (Default:1e-6)
Static Protected Member Functions
static void	dumpVar_int (CStream &out, const char *varName, int v)
	Used to print variable info from dumpToTextStream with the macro LOADABLEOPTS_DUMP_VAR.
static void	dumpVar_float (CStream &out, const char *varName, float v)
static void	dumpVar_double (CStream &out, const char *varName, double v)
static void	dumpVar_bool (CStream &out, const char *varName, bool v)
static void	dumpVar_string (CStream &out, const char *varName, const std::string &v)

Constructor & Destructor Documentation

mrpt::slam::CICP::TConfigParams::TConfigParams ( )

Initializer for default values:

Member Function Documentation

void mrpt::utils::CLoadableOptions::dumpToConsole ( ) const [inherited]

This method must display clearly all the contents of the structure in textual form, sending it to a CStream.

void mrpt::slam::CICP::TConfigParams::dumpToTextStream ( CStream & out ) const [virtual]

See utils::CLoadableOptions.

Implements mrpt::utils::CLoadableOptions.

static void mrpt::utils::CLoadableOptions::dumpVar_bool	(	CStream &	out,
		const char *	varName,
		bool	v
	)		`[static, protected, inherited]`

static void mrpt::utils::CLoadableOptions::dumpVar_double	(	CStream &	out,
		const char *	varName,
		double	v
	)		`[static, protected, inherited]`

static void mrpt::utils::CLoadableOptions::dumpVar_float	(	CStream &	out,
		const char *	varName,
		float	v
	)		`[static, protected, inherited]`

static void mrpt::utils::CLoadableOptions::dumpVar_int	(	CStream &	out,
		const char *	varName,
		int	v
	)		`[static, protected, inherited]`

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

static void mrpt::utils::CLoadableOptions::dumpVar_string	(	CStream &	out,
		const char *	varName,
		const std::string &	v
	)		`[static, protected, inherited]`

void mrpt::slam::CICP::TConfigParams::loadFromConfigFile	(	const mrpt::utils::CConfigFileBase &	source,
		const std::string &	section
	)		`[virtual]`

See utils::CLoadableOptions.

Implements mrpt::utils::CLoadableOptions.

void mrpt::utils::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

virtual void mrpt::utils::CLoadableOptions::saveToConfigFile	(	mrpt::utils::CConfigFileBase &	source,
		const std::string &	section
	)		`[inline, virtual, inherited]`

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

See also:: loadFromConfigFile, saveToConfigFileName

Reimplemented in mrpt::vision::TMultiResDescOptions, and mrpt::vision::TMultiResDescMatchOptions.

Definition at line 87 of file CLoadableOptions.h.

References THROW_EXCEPTION.

void mrpt::utils::CLoadableOptions::saveToConfigFileName	(	const std::string &	config_file,
		const std::string &	section
	)		`[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

Member Data Documentation

float mrpt::slam::CICP::TConfigParams::ALFA

The scale factor for threshold everytime convergence is achieved.

Definition at line 104 of file CICP.h.

float mrpt::slam::CICP::TConfigParams::Axy_aprox_derivatives

The size of the perturbance in x & y used to estimate the Jacobians of the square error (in LM & IKF methods, default=0.05).

Definition at line 147 of file CICP.h.

uint32_t mrpt::slam::CICP::TConfigParams::corresponding_points_decimation

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 159 of file CICP.h.

float mrpt::slam::CICP::TConfigParams::covariance_varPoints

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 the paper: ....