Point Cloud Library (PCL)  1.14.1-dev
List of all members | Public Types | Public Member Functions | Protected Member Functions | Protected Attributes
pcl::NormalDistributionsTransform2D< PointSource, PointTarget > Class Template Reference

NormalDistributionsTransform2D provides an implementation of the Normal Distributions Transform algorithm for scan matching. More...

#include <pcl/registration/ndt_2d.h>

+ Inheritance diagram for pcl::NormalDistributionsTransform2D< PointSource, PointTarget >:
+ Collaboration diagram for pcl::NormalDistributionsTransform2D< PointSource, PointTarget >:

Public Types

using Ptr = shared_ptr< NormalDistributionsTransform2D< PointSource, PointTarget > >
 
using ConstPtr = shared_ptr< const NormalDistributionsTransform2D< PointSource, PointTarget > >
 
- Public Types inherited from pcl::Registration< PointSource, PointTarget, Scalar >
using Matrix4 = Eigen::Matrix< Scalar, 4, 4 >
 
using Ptr = shared_ptr< Registration< PointSource, PointTarget, Scalar > >
 
using ConstPtr = shared_ptr< const Registration< PointSource, PointTarget, Scalar > >
 
using CorrespondenceRejectorPtr = pcl::registration::CorrespondenceRejector::Ptr
 
using KdTree = pcl::search::KdTree< PointTarget >
 
using KdTreePtr = typename KdTree::Ptr
 
using KdTreeReciprocal = pcl::search::KdTree< PointSource >
 
using KdTreeReciprocalPtr = typename KdTreeReciprocal::Ptr
 
using PointCloudSource = pcl::PointCloud< PointSource >
 
using PointCloudSourcePtr = typename PointCloudSource::Ptr
 
using PointCloudSourceConstPtr = typename PointCloudSource::ConstPtr
 
using PointCloudTarget = pcl::PointCloud< PointTarget >
 
using PointCloudTargetPtr = typename PointCloudTarget::Ptr
 
using PointCloudTargetConstPtr = typename PointCloudTarget::ConstPtr
 
using PointRepresentationConstPtr = typename KdTree::PointRepresentationConstPtr
 
using TransformationEstimation = typename pcl::registration::TransformationEstimation< PointSource, PointTarget, Scalar >
 
using TransformationEstimationPtr = typename TransformationEstimation::Ptr
 
using TransformationEstimationConstPtr = typename TransformationEstimation::ConstPtr
 
using CorrespondenceEstimation = pcl::registration::CorrespondenceEstimationBase< PointSource, PointTarget, Scalar >
 
using CorrespondenceEstimationPtr = typename CorrespondenceEstimation::Ptr
 
using CorrespondenceEstimationConstPtr = typename CorrespondenceEstimation::ConstPtr
 
using UpdateVisualizerCallbackSignature = void(const pcl::PointCloud< PointSource > &, const pcl::Indices &, const pcl::PointCloud< PointTarget > &, const pcl::Indices &)
 The callback signature to the function updating intermediate source point cloud position during it's registration to the target point cloud. More...
 
- Public Types inherited from pcl::PCLBase< PointSource >
using PointCloud = pcl::PointCloud< PointSource >
 
using PointCloudPtr = typename PointCloud::Ptr
 
using PointCloudConstPtr = typename PointCloud::ConstPtr
 
using PointIndicesPtr = PointIndices::Ptr
 
using PointIndicesConstPtr = PointIndices::ConstPtr
 

Public Member Functions

 NormalDistributionsTransform2D ()
 Empty constructor. More...
 
 ~NormalDistributionsTransform2D () override=default
 Empty destructor. More...
 
virtual void setGridCentre (const Eigen::Vector2f &centre)
 centre of the ndt grid (target coordinate system) More...
 
virtual void setGridStep (const Eigen::Vector2f &step)
 Grid spacing (step) of the NDT grid. More...
 
virtual void setGridExtent (const Eigen::Vector2f &extent)
 NDT Grid extent (in either direction from the grid centre) More...
 
virtual void setOptimizationStepSize (const double &lambda)
 NDT Newton optimisation step size parameter. More...
 
virtual void setOptimizationStepSize (const Eigen::Vector3d &lambda)
 NDT Newton optimisation step size parameter. More...
 
- Public Member Functions inherited from pcl::Registration< PointSource, PointTarget, Scalar >
 Registration ()
 Empty constructor. More...
 
 ~Registration () override=default
 destructor. More...
 
void setTransformationEstimation (const TransformationEstimationPtr &te)
 Provide a pointer to the transformation estimation object. More...
 
void setCorrespondenceEstimation (const CorrespondenceEstimationPtr &ce)
 Provide a pointer to the correspondence estimation object. More...
 
virtual void setInputSource (const PointCloudSourceConstPtr &cloud)
 Provide a pointer to the input source (e.g., the point cloud that we want to align to the target) More...
 
PointCloudSourceConstPtr const getInputSource ()
 Get a pointer to the input point cloud dataset target. More...
 
virtual void setInputTarget (const PointCloudTargetConstPtr &cloud)
 Provide a pointer to the input target (e.g., the point cloud that we want to align the input source to) More...
 
PointCloudTargetConstPtr const getInputTarget ()
 Get a pointer to the input point cloud dataset target. More...
 
void setSearchMethodTarget (const KdTreePtr &tree, bool force_no_recompute=false)
 Provide a pointer to the search object used to find correspondences in the target cloud. More...
 
KdTreePtr getSearchMethodTarget () const
 Get a pointer to the search method used to find correspondences in the target cloud. More...
 
void setSearchMethodSource (const KdTreeReciprocalPtr &tree, bool force_no_recompute=false)
 Provide a pointer to the search object used to find correspondences in the source cloud (usually used by reciprocal correspondence finding). More...
 
KdTreeReciprocalPtr getSearchMethodSource () const
 Get a pointer to the search method used to find correspondences in the source cloud. More...
 
Matrix4 getFinalTransformation ()
 Get the final transformation matrix estimated by the registration method. More...
 
Matrix4 getLastIncrementalTransformation ()
 Get the last incremental transformation matrix estimated by the registration method. More...
 
void setMaximumIterations (int nr_iterations)
 Set the maximum number of iterations the internal optimization should run for. More...
 
int getMaximumIterations ()
 Get the maximum number of iterations the internal optimization should run for, as set by the user. More...
 
void setRANSACIterations (int ransac_iterations)
 Set the number of iterations RANSAC should run for. More...
 
double getRANSACIterations ()
 Get the number of iterations RANSAC should run for, as set by the user. More...
 
void setRANSACOutlierRejectionThreshold (double inlier_threshold)
 Set the inlier distance threshold for the internal RANSAC outlier rejection loop. More...
 
double getRANSACOutlierRejectionThreshold ()
 Get the inlier distance threshold for the internal outlier rejection loop as set by the user. More...
 
void setMaxCorrespondenceDistance (double distance_threshold)
 Set the maximum distance threshold between two correspondent points in source <-> target. More...
 
double getMaxCorrespondenceDistance ()
 Get the maximum distance threshold between two correspondent points in source <-> target. More...
 
void setTransformationEpsilon (double epsilon)
 Set the transformation epsilon (maximum allowable translation squared difference between two consecutive transformations) in order for an optimization to be considered as having converged to the final solution. More...
 
double getTransformationEpsilon ()
 Get the transformation epsilon (maximum allowable translation squared difference between two consecutive transformations) as set by the user. More...
 
void setTransformationRotationEpsilon (double epsilon)
 Set the transformation rotation epsilon (maximum allowable rotation difference between two consecutive transformations) in order for an optimization to be considered as having converged to the final solution. More...
 
double getTransformationRotationEpsilon ()
 Get the transformation rotation epsilon (maximum allowable difference between two consecutive transformations) as set by the user (epsilon is the cos(angle) in a axis-angle representation). More...
 
void setEuclideanFitnessEpsilon (double epsilon)
 Set the maximum allowed Euclidean error between two consecutive steps in the ICP loop, before the algorithm is considered to have converged. More...
 
double getEuclideanFitnessEpsilon ()
 Get the maximum allowed distance error before the algorithm will be considered to have converged, as set by the user. More...
 
void setPointRepresentation (const PointRepresentationConstPtr &point_representation)
 Provide a boost shared pointer to the PointRepresentation to be used when comparing points. More...
 
bool registerVisualizationCallback (std::function< UpdateVisualizerCallbackSignature > &visualizerCallback)
 Register the user callback function which will be called from registration thread in order to update point cloud obtained after each iteration. More...
 
double getFitnessScore (double max_range=std::numeric_limits< double >::max())
 Obtain the Euclidean fitness score (e.g., mean of squared distances from the source to the target) More...
 
double getFitnessScore (const std::vector< float > &distances_a, const std::vector< float > &distances_b)
 Obtain the Euclidean fitness score (e.g., mean of squared distances from the source to the target) from two sets of correspondence distances (distances between source and target points) More...
 
bool hasConverged () const
 Return the state of convergence after the last align run. More...
 
void align (PointCloudSource &output)
 Call the registration algorithm which estimates the transformation and returns the transformed source (input) as output. More...
 
void align (PointCloudSource &output, const Matrix4 &guess)
 Call the registration algorithm which estimates the transformation and returns the transformed source (input) as output. More...
 
const std::string & getClassName () const
 Abstract class get name method. More...
 
bool initCompute ()
 Internal computation initialization. More...
 
bool initComputeReciprocal ()
 Internal computation when reciprocal lookup is needed. More...
 
void addCorrespondenceRejector (const CorrespondenceRejectorPtr &rejector)
 Add a new correspondence rejector to the list. More...
 
std::vector< CorrespondenceRejectorPtrgetCorrespondenceRejectors ()
 Get the list of correspondence rejectors. More...
 
bool removeCorrespondenceRejector (unsigned int i)
 Remove the i-th correspondence rejector in the list. More...
 
void clearCorrespondenceRejectors ()
 Clear the list of correspondence rejectors. More...
 
- Public Member Functions inherited from pcl::PCLBase< PointSource >
 PCLBase ()
 Empty constructor. More...
 
 PCLBase (const PCLBase &base)
 Copy constructor. More...
 
virtual ~PCLBase ()=default
 Destructor. More...
 
virtual void setInputCloud (const PointCloudConstPtr &cloud)
 Provide a pointer to the input dataset. More...
 
PointCloudConstPtr const getInputCloud () const
 Get a pointer to the input point cloud dataset. More...
 
virtual void setIndices (const IndicesPtr &indices)
 Provide a pointer to the vector of indices that represents the input data. More...
 
virtual void setIndices (const IndicesConstPtr &indices)
 Provide a pointer to the vector of indices that represents the input data. More...
 
virtual void setIndices (const PointIndicesConstPtr &indices)
 Provide a pointer to the vector of indices that represents the input data. More...
 
virtual void setIndices (std::size_t row_start, std::size_t col_start, std::size_t nb_rows, std::size_t nb_cols)
 Set the indices for the points laying within an interest region of the point cloud. More...
 
IndicesPtr getIndices ()
 Get a pointer to the vector of indices used. More...
 
IndicesConstPtr const getIndices () const
 Get a pointer to the vector of indices used. More...
 
const PointSource & operator[] (std::size_t pos) const
 Override PointCloud operator[] to shorten code. More...
 

Protected Member Functions

void computeTransformation (PointCloudSource &output, const Eigen::Matrix4f &guess) override
 Rigid transformation computation method with initial guess. More...
 
- Protected Member Functions inherited from pcl::Registration< PointSource, PointTarget, Scalar >
bool searchForNeighbors (const PointCloudSource &cloud, int index, pcl::Indices &indices, std::vector< float > &distances)
 Search for the closest nearest neighbor of a given point. More...
 
virtual void computeTransformation (PointCloudSource &output, const Matrix4 &guess)=0
 Abstract transformation computation method with initial guess. More...
 
- Protected Member Functions inherited from pcl::PCLBase< PointSource >
bool initCompute ()
 This method should get called before starting the actual computation. More...
 
bool deinitCompute ()
 This method should get called after finishing the actual computation. More...
 

Protected Attributes

Eigen::Vector2f grid_centre_
 
Eigen::Vector2f grid_step_
 
Eigen::Vector2f grid_extent_
 
Eigen::Vector3d newton_lambda_
 
- Protected Attributes inherited from pcl::Registration< PointSource, PointTarget, Scalar >
std::string reg_name_
 The registration method name. More...
 
KdTreePtr tree_
 A pointer to the spatial search object. More...
 
KdTreeReciprocalPtr tree_reciprocal_
 A pointer to the spatial search object of the source. More...
 
int nr_iterations_ {0}
 The number of iterations the internal optimization ran for (used internally). More...
 
int max_iterations_ {10}
 The maximum number of iterations the internal optimization should run for. More...
 
int ransac_iterations_ {0}
 The number of iterations RANSAC should run for. More...
 
PointCloudTargetConstPtr target_
 The input point cloud dataset target. More...
 
Matrix4 final_transformation_
 The final transformation matrix estimated by the registration method after N iterations. More...
 
Matrix4 transformation_
 The transformation matrix estimated by the registration method. More...
 
Matrix4 previous_transformation_
 The previous transformation matrix estimated by the registration method (used internally). More...
 
double transformation_epsilon_ {0.0}
 The maximum difference between two consecutive transformations in order to consider convergence (user defined). More...
 
double transformation_rotation_epsilon_ {0.0}
 The maximum rotation difference between two consecutive transformations in order to consider convergence (user defined). More...
 
double euclidean_fitness_epsilon_
 The maximum allowed Euclidean error between two consecutive steps in the ICP loop, before the algorithm is considered to have converged. More...
 
double corr_dist_threshold_
 The maximum distance threshold between two correspondent points in source <-> target. More...
 
double inlier_threshold_ {0.05}
 The inlier distance threshold for the internal RANSAC outlier rejection loop. More...
 
bool converged_ {false}
 Holds internal convergence state, given user parameters. More...
 
unsigned int min_number_correspondences_ {3}
 The minimum number of correspondences that the algorithm needs before attempting to estimate the transformation. More...
 
CorrespondencesPtr correspondences_
 The set of correspondences determined at this ICP step. More...
 
TransformationEstimationPtr transformation_estimation_
 A TransformationEstimation object, used to calculate the 4x4 rigid transformation. More...
 
CorrespondenceEstimationPtr correspondence_estimation_
 A CorrespondenceEstimation object, used to estimate correspondences between the source and the target cloud. More...
 
std::vector< CorrespondenceRejectorPtrcorrespondence_rejectors_
 The list of correspondence rejectors to use. More...
 
bool target_cloud_updated_ {true}
 Variable that stores whether we have a new target cloud, meaning we need to pre-process it again. More...
 
bool source_cloud_updated_ {true}
 Variable that stores whether we have a new source cloud, meaning we need to pre-process it again. More...
 
bool force_no_recompute_ {false}
 A flag which, if set, means the tree operating on the target cloud will never be recomputed. More...
 
bool force_no_recompute_reciprocal_ {false}
 A flag which, if set, means the tree operating on the source cloud will never be recomputed. More...
 
std::function< UpdateVisualizerCallbackSignatureupdate_visualizer_
 Callback function to update intermediate source point cloud position during it's registration to the target point cloud. More...
 
- Protected Attributes inherited from pcl::PCLBase< PointSource >
PointCloudConstPtr input_
 The input point cloud dataset. More...
 
IndicesPtr indices_
 A pointer to the vector of point indices to use. More...
 
bool use_indices_
 Set to true if point indices are used. More...
 
bool fake_indices_
 If no set of indices are given, we construct a set of fake indices that mimic the input PointCloud. More...
 

Detailed Description

template<typename PointSource, typename PointTarget>
class pcl::NormalDistributionsTransform2D< PointSource, PointTarget >

NormalDistributionsTransform2D provides an implementation of the Normal Distributions Transform algorithm for scan matching.

This implementation is intended to match the definition: Peter Biber and Wolfgang Straßer. The normal distributions transform: A new approach to laser scan matching. In Proceedings of the IEEE In- ternational Conference on Intelligent Robots and Systems (IROS), pages 2743–2748, Las Vegas, USA, October 2003.

Author
James Crosby

Definition at line 60 of file ndt_2d.h.

Member Typedef Documentation

◆ ConstPtr

template<typename PointSource , typename PointTarget >
using pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::ConstPtr = shared_ptr<const NormalDistributionsTransform2D<PointSource, PointTarget> >

Definition at line 74 of file ndt_2d.h.

◆ Ptr

template<typename PointSource , typename PointTarget >
using pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::Ptr = shared_ptr<NormalDistributionsTransform2D<PointSource, PointTarget> >

Definition at line 73 of file ndt_2d.h.

Constructor & Destructor Documentation

◆ NormalDistributionsTransform2D()

template<typename PointSource , typename PointTarget >
pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::NormalDistributionsTransform2D ( )
inline

Empty constructor.

Definition at line 78 of file ndt_2d.h.

References pcl::Registration< PointSource, PointTarget, Scalar >::reg_name_.

◆ ~NormalDistributionsTransform2D()

template<typename PointSource , typename PointTarget >
pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::~NormalDistributionsTransform2D ( )
overridedefault

Empty destructor.

Member Function Documentation

◆ computeTransformation()

template<typename PointSource , typename PointTarget >
void pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::computeTransformation ( PointCloudSource &  output,
const Eigen::Matrix4f &  guess 
)
overrideprotected

Rigid transformation computation method with initial guess.

Parameters
[out]outputthe transformed input point cloud dataset using the rigid transformation found
[in]guessthe initial guess of the transformation to compute

Definition at line 392 of file ndt_2d.hpp.

References pcl::ndt2d::ValueAndDerivatives< N, T >::grad, pcl::ndt2d::ValueAndDerivatives< N, T >::hessian, pcl::ndt2d::NDT2D< PointT >::test(), pcl::transformPointCloud(), pcl::ndt2d::ValueAndDerivatives< N, T >::value, and pcl::ndt2d::ValueAndDerivatives< N, T >::Zero().

◆ setGridCentre()

template<typename PointSource , typename PointTarget >
virtual void pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::setGridCentre ( const Eigen::Vector2f &  centre)
inlinevirtual

centre of the ndt grid (target coordinate system)

Parameters
centrevalue to set

Definition at line 95 of file ndt_2d.h.

References pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::grid_centre_.

◆ setGridExtent()

template<typename PointSource , typename PointTarget >
virtual void pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::setGridExtent ( const Eigen::Vector2f &  extent)
inlinevirtual

NDT Grid extent (in either direction from the grid centre)

Parameters
[in]extentvalue to set

Definition at line 113 of file ndt_2d.h.

References pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::grid_extent_.

◆ setGridStep()

template<typename PointSource , typename PointTarget >
virtual void pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::setGridStep ( const Eigen::Vector2f &  step)
inlinevirtual

Grid spacing (step) of the NDT grid.

Parameters
[in]stepvalue to set

Definition at line 104 of file ndt_2d.h.

References pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::grid_step_.

◆ setOptimizationStepSize() [1/2]

template<typename PointSource , typename PointTarget >
virtual void pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::setOptimizationStepSize ( const double &  lambda)
inlinevirtual

NDT Newton optimisation step size parameter.

Parameters
[in]lambdastep size: 1 is simple newton optimisation, smaller values may improve convergence

Definition at line 123 of file ndt_2d.h.

References pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::newton_lambda_.

◆ setOptimizationStepSize() [2/2]

template<typename PointSource , typename PointTarget >
virtual void pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::setOptimizationStepSize ( const Eigen::Vector3d &  lambda)
inlinevirtual

NDT Newton optimisation step size parameter.

Parameters
[in]lambdastep size: (1,1,1) is simple newton optimisation, smaller values may improve convergence, or elements may be set to zero to prevent optimisation over some parameters

This overload allows control of updates to the individual (x, y, theta) free parameters in the optimisation. If, for example, theta is believed to be close to the correct value a small value of lambda[2] should be used.

Definition at line 139 of file ndt_2d.h.

References pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::newton_lambda_.

Member Data Documentation

◆ grid_centre_

template<typename PointSource , typename PointTarget >
Eigen::Vector2f pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::grid_centre_
protected

◆ grid_extent_

template<typename PointSource , typename PointTarget >
Eigen::Vector2f pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::grid_extent_
protected

◆ grid_step_

template<typename PointSource , typename PointTarget >
Eigen::Vector2f pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::grid_step_
protected

◆ newton_lambda_

template<typename PointSource , typename PointTarget >
Eigen::Vector3d pcl::NormalDistributionsTransform2D< PointSource, PointTarget >::newton_lambda_
protected

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