MRPT logo

mrpt::slam::CGridMapAligner::TConfigParams Class Reference

The ICP algorithm configuration data. More...

#include <mrpt/slam/CGridMapAligner.h>

Inheritance diagram for mrpt::slam::CGridMapAligner::TConfigParams:

mrpt::utils::CLoadableOptions

List of all members.

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.

Public Attributes

TAlignerMethod methodSelection
 The aligner method:.
mrpt::vision::TDescriptorType feature_descriptor
 The feature descriptor to use: 0=detector already has descriptor, 1= SIFT, 2=SURF, 4=Spin images, 8=Polar images, 16=log-polar images.
mrpt::vision::CFeatureExtraction::TOptions feature_detector_options
 All the parameters for the feature detector.
float ransac_minSetSizeRatio
 RANSAC-step options:.
float ransac_SOG_sigma_m
 The square root of the uncertainty normalization variance var_m (see papers...).
float ransac_mahalanobisDistanceThreshold
 [amRobustMatch method only] RANSAC-step options:
double ransac_chi2_quantile
 [amModifiedRANSAC method only] The quantile used for chi-square thresholding (default=0.99)
double ransac_prob_good_inliers
 Probability of having a good inliers (def:0,9999), used for automatic number of iterations.
float featsPerSquareMeter
 Features extraction from grid map: How many features to extract.
float threshold_max
 Correspondences are considered if their distances are below this threshold (in the range [0,1]) (default=0.15).
float threshold_delta
 Correspondences are considered if their distances to the best match are below this threshold (in the range [0,1]) (default=0.15).
float min_ICP_goodness
 The minimum goodness (0-1) of the post-matching ICP to accept a hypothesis as good (default=0.25).
double max_ICP_mahadist
 The maximum Mahalanobis distance between the initial and final poses in the ICP not to discard the hypothesis (default=10).
double maxKLd_for_merge
 Maximum KL-divergence for merging modes of the SOG (default=0.9).
bool save_feat_coors
 DEBUG - Dump all feature correspondences in a directory "grid_feats".
bool debug_show_corrs
 DEBUG - Show graphs with the details of each feature correspondences.
bool debug_save_map_pairs
 DEBUG - Save the pair of maps with all the pairings.


Detailed Description

The ICP algorithm configuration data.

Definition at line 92 of file CGridMapAligner.h.


Constructor & Destructor Documentation

mrpt::slam::CGridMapAligner::TConfigParams::TConfigParams (  ) 

Initializer for default values:.


Member Function Documentation

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

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


Member Data Documentation

DEBUG - Save the pair of maps with all the pairings.

Definition at line 142 of file CGridMapAligner.h.

DEBUG - Show graphs with the details of each feature correspondences.

Definition at line 141 of file CGridMapAligner.h.

Features extraction from grid map: How many features to extract.

Definition at line 132 of file CGridMapAligner.h.

The feature descriptor to use: 0=detector already has descriptor, 1= SIFT, 2=SURF, 4=Spin images, 8=Polar images, 16=log-polar images.

Definition at line 113 of file CGridMapAligner.h.

All the parameters for the feature detector.

Definition at line 116 of file CGridMapAligner.h.

The maximum Mahalanobis distance between the initial and final poses in the ICP not to discard the hypothesis (default=10).

Definition at line 137 of file CGridMapAligner.h.

Maximum KL-divergence for merging modes of the SOG (default=0.9).

Definition at line 138 of file CGridMapAligner.h.

The aligner method:.

Definition at line 110 of file CGridMapAligner.h.

The minimum goodness (0-1) of the post-matching ICP to accept a hypothesis as good (default=0.25).

Definition at line 136 of file CGridMapAligner.h.

[amModifiedRANSAC method only] The quantile used for chi-square thresholding (default=0.99)

Definition at line 129 of file CGridMapAligner.h.

[amRobustMatch method only] RANSAC-step options:

See also:
CICP::robustRigidTransformation

Definition at line 127 of file CGridMapAligner.h.

RANSAC-step options:.

See also:
CICP::robustRigidTransformation The ratio of landmarks that must be inliers to accepto an hypotheses (typ: 0.20)

Definition at line 121 of file CGridMapAligner.h.

Probability of having a good inliers (def:0,9999), used for automatic number of iterations.

Definition at line 131 of file CGridMapAligner.h.

The square root of the uncertainty normalization variance var_m (see papers...).

Definition at line 122 of file CGridMapAligner.h.

DEBUG - Dump all feature correspondences in a directory "grid_feats".

Definition at line 140 of file CGridMapAligner.h.

Correspondences are considered if their distances to the best match are below this threshold (in the range [0,1]) (default=0.15).

Definition at line 134 of file CGridMapAligner.h.

Correspondences are considered if their distances are below this threshold (in the range [0,1]) (default=0.15).

Definition at line 133 of file CGridMapAligner.h.




Page generated by Doxygen 1.5.9 for MRPT 0.7.1 SVN: at Mon Aug 17 22:20:53 EDT 2009