MRPT  2.0.2
List of all members | Classes | Public Types | Public Member Functions | Static Public Member Functions | Static Public Attributes | Protected Member Functions | Protected Attributes
mrpt::poses::CPosePDFSOG Class Referenceabstract

Detailed Description

Declares a class that represents a Probability Density function (PDF) of a 2D pose $ p(\mathbf{x}) = [x ~ y ~ \phi ]^t $.

This class implements that PDF as the following multi-modal Gaussian distribution:

$ p(\mathbf{x}) = \sum\limits_{i=1}^N \omega^i \mathcal{N}( \mathbf{x} ; \bar{\mathbf{x}}^i, \mathbf{\Sigma}^i ) $

Where the number of modes N is the size of CPosePDFSOG::m_modes

See mrpt::poses::CPosePDF for more details.

See also
CPose2D, CPosePDF, CPosePDFParticles

Definition at line 34 of file CPosePDFSOG.h.

#include <mrpt/poses/CPosePDFSOG.h>

Inheritance diagram for mrpt::poses::CPosePDFSOG:

Classes

struct  TGaussianMode
 The struct for each mode: More...
 

Public Types

using CListGaussianModes = std::vector< TGaussianMode >
 
using const_iterator = CListGaussianModes::const_iterator
 
using iterator = CListGaussianModes::iterator
 
enum  { is_3D_val = 0 }
 
enum  { is_PDF_val = 1 }
 
using type_value = CPose2D
 The type of the state the PDF represents. More...
 
using self_t = CProbabilityDensityFunction< CPose2D, STATE_LEN >
 
using cov_mat_t = mrpt::math::CMatrixFixed< double, STATE_LEN, STATE_LEN >
 Covariance matrix type. More...
 
using inf_mat_t = cov_mat_t
 Information matrix type. More...
 

Public Member Functions

const CListGaussianModesgetSOGModes () const
 
 CPosePDFSOG (size_t nModes=1)
 Default constructor. More...
 
size_t size () const
 Return the number of Gaussian modes. More...
 
bool empty () const
 Return whether there is any Gaussian mode. More...
 
void clear ()
 Clear the list of modes. More...
 
const TGaussianModeoperator[] (size_t i) const
 Access to individual beacons. More...
 
TGaussianModeoperator[] (size_t i)
 Access to individual beacons. More...
 
const TGaussianModeget (size_t i) const
 Access to individual beacons. More...
 
TGaussianModeget (size_t i)
 Access to individual beacons. More...
 
void push_back (const TGaussianMode &m)
 Inserts a copy of the given mode into the SOG. More...
 
iterator begin ()
 
iterator end ()
 
const_iterator begin () const
 
const_iterator end () const
 
iterator erase (iterator i)
 
void resize (const size_t N)
 Resize the number of SOG modes. More...
 
void mergeModes (double max_KLd=0.5, bool verbose=false)
 Merge very close modes so the overall number of modes is reduced while preserving the total distribution. More...
 
void getMean (CPose2D &mean_pose) const override
 
std::tuple< cov_mat_t, type_valuegetCovarianceAndMean () const override
 Returns an estimate of the pose covariance matrix (STATE_LENxSTATE_LEN cov matrix) and the mean, both at once. More...
 
void getMostLikelyCovarianceAndMean (mrpt::math::CMatrixDouble33 &cov, CPose2D &mean_point) const
 For the most likely Gaussian mode in the SOG, returns the pose covariance matrix (3x3 cov matrix) and the mean. More...
 
void normalizeWeights ()
 Normalize the weights in m_modes such as the maximum log-weight is 0. More...
 
void copyFrom (const CPosePDF &o) override
 Copy operator, translating if necesary (for example, between particles and gaussian representations) More...
 
bool saveToTextFile (const std::string &file) const override
 Save the density to a text file, with the following format: There is one row per Gaussian "mode", and each row contains 10 elements: More...
 
void changeCoordinatesReference (const CPose3D &newReferenceBase) override
 this = p (+) this. More...
 
void rotateAllCovariances (double ang)
 Rotate all the covariance matrixes by replacing them by $ \mathbf{R}~\mathbf{COV}~\mathbf{R}^t $, where $ \mathbf{R} = \left[ \begin{array}{ccc} \cos\alpha & -\sin\alpha & 0 \\ \sin\alpha & \cos\alpha & 0 \\ 0 & 0 & 1 \end{array}\right] $. More...
 
void drawSingleSample (CPose2D &outPart) const override
 Draws a single sample from the distribution. More...
 
void drawManySamples (size_t N, std::vector< mrpt::math::CVectorDouble > &outSamples) const override
 Draws a number of samples from the distribution, and saves as a list of 1x3 vectors, where each row contains a (x,y,phi) datum. More...
 
void inverse (CPosePDF &o) const override
 Returns a new PDF such as: NEW_PDF = (0,0,0) - THIS_PDF. More...
 
void operator+= (const mrpt::poses::CPose2D &Ap)
 Makes: thisPDF = thisPDF + Ap, where "+" is pose composition (both the mean, and the covariance matrix are updated). More...
 
double evaluatePDF (const mrpt::poses::CPose2D &x, bool sumOverAllPhis=false) const
 Evaluates the PDF at a given point. More...
 
double evaluateNormalizedPDF (const mrpt::poses::CPose2D &x) const
 Evaluates the ratio PDF(x) / max_PDF(x*), that is, the normalized PDF in the range [0,1]. More...
 
void evaluatePDFInArea (double x_min, double x_max, double y_min, double y_max, double resolutionXY, double phi, mrpt::math::CMatrixDouble &outMatrix, bool sumOverAllPhis=false)
 Evaluates the PDF within a rectangular grid (and a fixed orientation) and saves the result in a matrix (each row contains values for a fixed y-coordinate value). More...
 
void bayesianFusion (const CPosePDF &p1, const CPosePDF &p2, const double minMahalanobisDistToDrop=0) override
 Bayesian fusion of two pose distributions, then save the result in this object (WARNING: Currently p1 must be a mrpt::poses::CPosePDFSOG object and p2 a mrpt::poses::CPosePDFGaussian object) More...
 
template<class OPENGL_SETOFOBJECTSPTR >
void getAs3DObject (OPENGL_SETOFOBJECTSPTR &out_obj) const
 Returns a 3D representation of this PDF (it doesn't clear the current contents of out_obj, but append new OpenGL objects to that list) More...
 
template<class OPENGL_SETOFOBJECTSPTR >
OPENGL_SETOFOBJECTSPTR getAs3DObject () const
 Returns a 3D representation of this PDF. More...
 
virtual mxArraywriteToMatlab () const
 Introduces a pure virtual method responsible for writing to a mxArray Matlab object, typically a MATLAB struct whose contents are documented in each derived class. More...
 
virtual void getMean (type_value &mean_point) const=0
 Returns the mean, or mathematical expectation of the probability density distribution (PDF). More...
 
virtual void getCovarianceAndMean (cov_mat_t &c, CPose2D &mean) const final
 
void getCovarianceDynAndMean (mrpt::math::CMatrixDouble &cov, type_value &mean_point) const
 Returns an estimate of the pose covariance matrix (STATE_LENxSTATE_LEN cov matrix) and the mean, both at once. More...
 
type_value getMeanVal () const
 Returns the mean, or mathematical expectation of the probability density distribution (PDF). More...
 
void getCovariance (mrpt::math::CMatrixDouble &cov) const
 Returns the estimate of the covariance matrix (STATE_LEN x STATE_LEN covariance matrix) More...
 
void getCovariance (cov_mat_t &cov) const
 Returns the estimate of the covariance matrix (STATE_LEN x STATE_LEN covariance matrix) More...
 
cov_mat_t getCovariance () const
 Returns the estimate of the covariance matrix (STATE_LEN x STATE_LEN covariance matrix) More...
 
virtual bool isInfType () const
 Returns whether the class instance holds the uncertainty in covariance or information form. More...
 
virtual void getInformationMatrix (inf_mat_t &inf) const
 Returns the information (inverse covariance) matrix (a STATE_LEN x STATE_LEN matrix) Unless reimplemented in derived classes, this method first reads the covariance, then invert it. More...
 
virtual void drawSingleSample (CPose2D &outPart) const=0
 Draws a single sample from the distribution. More...
 
double getCovarianceEntropy () const
 Compute the entropy of the estimated covariance matrix. More...
 
RTTI classes and functions for polymorphic hierarchies
mrpt::rtti::CObject::Ptr duplicateGetSmartPtr () const
 Makes a deep copy of the object and returns a smart pointer to it. More...
 

Static Public Member Functions

static void jacobiansPoseComposition (const CPose2D &x, const CPose2D &u, mrpt::math::CMatrixDouble33 &df_dx, mrpt::math::CMatrixDouble33 &df_du, const bool compute_df_dx=true, const bool compute_df_du=true)
 This static method computes the pose composition Jacobians, with these formulas: More...
 
static void jacobiansPoseComposition (const CPosePDFGaussian &x, const CPosePDFGaussian &u, mrpt::math::CMatrixDouble33 &df_dx, mrpt::math::CMatrixDouble33 &df_du)
 
static constexpr bool is_3D ()
 
static constexpr bool is_PDF ()
 

Static Public Attributes

static constexpr size_t state_length
 The length of the variable, for example, 3 for a 3D point, 6 for a 3D pose (x y z yaw pitch roll). More...
 

Protected Member Functions

void enforceCovSymmetry ()
 Ensures the symmetry of the covariance matrix (eventually certain operations in the math-coprocessor lead to non-symmetric matrixes!) More...
 
CSerializable virtual methods
uint8_t serializeGetVersion () const override
 Must return the current versioning number of the object. More...
 
void serializeTo (mrpt::serialization::CArchive &out) const override
 Pure virtual method for writing (serializing) to an abstract archive. More...
 
void serializeFrom (mrpt::serialization::CArchive &in, uint8_t serial_version) override
 Pure virtual method for reading (deserializing) from an abstract archive. More...
 
CSerializable virtual methods
virtual void serializeTo (CSchemeArchiveBase &out) const
 Virtual method for writing (serializing) to an abstract schema based archive. More...
 
virtual void serializeFrom (CSchemeArchiveBase &in)
 Virtual method for reading (deserializing) from an abstract schema based archive. More...
 

Protected Attributes

CListGaussianModes m_modes
 The list of SOG modes. More...
 

RTTI stuff

using Ptr = std::shared_ptr< mrpt::poses ::CPosePDFSOG >
 
using ConstPtr = std::shared_ptr< const mrpt::poses ::CPosePDFSOG >
 
using UniquePtr = std::unique_ptr< mrpt::poses ::CPosePDFSOG >
 
using ConstUniquePtr = std::unique_ptr< const mrpt::poses ::CPosePDFSOG >
 
static const mrpt::rtti::TRuntimeClassId runtimeClassId
 
static constexpr const char * className = "mrpt::poses" "::" "CPosePDFSOG"
 
static const mrpt::rtti::TRuntimeClassId_GetBaseClass ()
 
static constexpr auto getClassName ()
 
static const mrpt::rtti::TRuntimeClassIdGetRuntimeClassIdStatic ()
 
static std::shared_ptr< CObjectCreateObject ()
 
template<typename... Args>
static Ptr Create (Args &&... args)
 
template<typename Alloc , typename... Args>
static Ptr CreateAlloc (const Alloc &alloc, Args &&... args)
 
template<typename... Args>
static UniquePtr CreateUnique (Args &&... args)
 
virtual const mrpt::rtti::TRuntimeClassIdGetRuntimeClass () const override
 Returns information about the class of an object in runtime. More...
 
virtual mrpt::rtti::CObjectclone () const override
 Returns a deep copy (clone) of the object, indepently of its class. More...
 

Member Typedef Documentation

◆ CListGaussianModes

Definition at line 65 of file CPosePDFSOG.h.

◆ const_iterator

using mrpt::poses::CPosePDFSOG::const_iterator = CListGaussianModes::const_iterator

Definition at line 66 of file CPosePDFSOG.h.

◆ ConstPtr

Definition at line 36 of file CPosePDFSOG.h.

◆ ConstUniquePtr

using mrpt::poses::CPosePDFSOG::ConstUniquePtr = std::unique_ptr<const mrpt::poses :: CPosePDFSOG >

Definition at line 36 of file CPosePDFSOG.h.

◆ cov_mat_t

using mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::cov_mat_t = mrpt::math::CMatrixFixed<double, STATE_LEN, STATE_LEN>
inherited

Covariance matrix type.

Definition at line 37 of file CProbabilityDensityFunction.h.

◆ inf_mat_t

Information matrix type.

Definition at line 39 of file CProbabilityDensityFunction.h.

◆ iterator

using mrpt::poses::CPosePDFSOG::iterator = CListGaussianModes::iterator

Definition at line 67 of file CPosePDFSOG.h.

◆ Ptr

A type for the associated smart pointer

Definition at line 36 of file CPosePDFSOG.h.

◆ self_t

using mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::self_t = CProbabilityDensityFunction<CPose2D , STATE_LEN>
inherited

Definition at line 35 of file CProbabilityDensityFunction.h.

◆ type_value

The type of the state the PDF represents.

Definition at line 34 of file CProbabilityDensityFunction.h.

◆ UniquePtr

using mrpt::poses::CPosePDFSOG::UniquePtr = std::unique_ptr< mrpt::poses :: CPosePDFSOG >

Definition at line 36 of file CPosePDFSOG.h.

Member Enumeration Documentation

◆ anonymous enum

anonymous enum
inherited
Enumerator
is_3D_val 

Definition at line 94 of file CPosePDF.h.

◆ anonymous enum

anonymous enum
inherited
Enumerator
is_PDF_val 

Definition at line 99 of file CPosePDF.h.

Constructor & Destructor Documentation

◆ CPosePDFSOG()

CPosePDFSOG::CPosePDFSOG ( size_t  nModes = 1)

Default constructor.

Parameters
nModesThe initial size of CPosePDFSOG::m_modes

Definition at line 36 of file CPosePDFSOG.cpp.

Member Function Documentation

◆ _GetBaseClass()

static const mrpt::rtti::TRuntimeClassId* mrpt::poses::CPosePDFSOG::_GetBaseClass ( )
staticprotected

◆ bayesianFusion()

void CPosePDFSOG::bayesianFusion ( const CPosePDF p1,
const CPosePDF p2,
const double  minMahalanobisDistToDrop = 0 
)
overridevirtual

◆ begin() [1/2]

iterator mrpt::poses::CPosePDFSOG::begin ( )
inline

Definition at line 119 of file CPosePDFSOG.h.

References m_modes.

◆ begin() [2/2]

const_iterator mrpt::poses::CPosePDFSOG::begin ( ) const
inline

Definition at line 121 of file CPosePDFSOG.h.

References m_modes.

◆ changeCoordinatesReference()

void CPosePDFSOG::changeCoordinatesReference ( const CPose3D newReferenceBase)
overridevirtual

this = p (+) this.

This can be used to convert a PDF from local coordinates to global, providing the point (newReferenceBase) from which "to project" the current pdf. Result PDF substituted the currently stored one in the object.

Implements mrpt::poses::CPosePDF.

Definition at line 199 of file CPosePDFSOG.cpp.

References mrpt::math::MatrixVectorBase< Scalar, Derived >::block(), mrpt::poses::CPose2D::getHomogeneousMatrix(), and mrpt::math::multiply_HCHt().

Here is the call graph for this function:

◆ clear()

void CPosePDFSOG::clear ( )

Clear the list of modes.

Definition at line 40 of file CPosePDFSOG.cpp.

◆ clone()

virtual mrpt::rtti::CObject* mrpt::poses::CPosePDFSOG::clone ( ) const
overridevirtual

Returns a deep copy (clone) of the object, indepently of its class.

Implements mrpt::rtti::CObject.

◆ copyFrom()

void CPosePDFSOG::copyFrom ( const CPosePDF o)
overridevirtual

Copy operator, translating if necesary (for example, between particles and gaussian representations)

Implements mrpt::poses::CPosePDF.

Definition at line 157 of file CPosePDFSOG.cpp.

References CLASS_ID, mrpt::math::cov(), mrpt::math::CProbabilityDensityFunction< TDATA, STATE_LEN >::getCovariance(), mrpt::math::CProbabilityDensityFunction< TDATA, STATE_LEN >::getMean(), mrpt::poses::CPosePDF::GetRuntimeClass(), mrpt::math::mean(), MRPT_END, MRPT_START, and resize().

Here is the call graph for this function:

◆ Create()

template<typename... Args>
static Ptr mrpt::poses::CPosePDFSOG::Create ( Args &&...  args)
inlinestatic

Definition at line 36 of file CPosePDFSOG.h.

◆ CreateAlloc()

template<typename Alloc , typename... Args>
static Ptr mrpt::poses::CPosePDFSOG::CreateAlloc ( const Alloc &  alloc,
Args &&...  args 
)
inlinestatic

Definition at line 36 of file CPosePDFSOG.h.

◆ CreateObject()

static std::shared_ptr<CObject> mrpt::poses::CPosePDFSOG::CreateObject ( )
static

◆ CreateUnique()

template<typename... Args>
static UniquePtr mrpt::poses::CPosePDFSOG::CreateUnique ( Args &&...  args)
inlinestatic

Definition at line 36 of file CPosePDFSOG.h.

◆ drawManySamples()

void CPosePDFSOG::drawManySamples ( size_t  N,
std::vector< mrpt::math::CVectorDouble > &  outSamples 
) const
overridevirtual

Draws a number of samples from the distribution, and saves as a list of 1x3 vectors, where each row contains a (x,y,phi) datum.

Reimplemented from mrpt::math::CProbabilityDensityFunction< CPose2D, 3 >.

Definition at line 258 of file CPosePDFSOG.cpp.

References MRPT_END, MRPT_START, and THROW_EXCEPTION.

◆ drawSingleSample() [1/2]

virtual void mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::drawSingleSample ( CPose2D &  outPart) const
pure virtualinherited

Draws a single sample from the distribution.

◆ drawSingleSample() [2/2]

void CPosePDFSOG::drawSingleSample ( CPose2D outPart) const
override

Draws a single sample from the distribution.

Definition at line 248 of file CPosePDFSOG.cpp.

References MRPT_END, MRPT_START, and THROW_EXCEPTION.

◆ duplicateGetSmartPtr()

mrpt::rtti::CObject::Ptr CObject::duplicateGetSmartPtr ( ) const
inlineinherited

Makes a deep copy of the object and returns a smart pointer to it.

Definition at line 204 of file CObject.h.

References mrpt::rtti::CObject::clone().

Referenced by mrpt::obs::CRawlog::insert().

Here is the call graph for this function:
Here is the caller graph for this function:

◆ empty()

bool mrpt::poses::CPosePDFSOG::empty ( ) const
inline

Return whether there is any Gaussian mode.

Definition at line 87 of file CPosePDFSOG.h.

References m_modes.

◆ end() [1/2]

iterator mrpt::poses::CPosePDFSOG::end ( )
inline

Definition at line 120 of file CPosePDFSOG.h.

References m_modes.

◆ end() [2/2]

const_iterator mrpt::poses::CPosePDFSOG::end ( ) const
inline

Definition at line 122 of file CPosePDFSOG.h.

References m_modes.

◆ enforceCovSymmetry()

void CPosePDFSOG::enforceCovSymmetry ( )
protected

Ensures the symmetry of the covariance matrix (eventually certain operations in the math-coprocessor lead to non-symmetric matrixes!)

Definition at line 453 of file CPosePDFSOG.cpp.

◆ erase()

iterator mrpt::poses::CPosePDFSOG::erase ( iterator  i)
inline

Definition at line 123 of file CPosePDFSOG.h.

References m_modes.

◆ evaluateNormalizedPDF()

double CPosePDFSOG::evaluateNormalizedPDF ( const mrpt::poses::CPose2D x) const

Evaluates the ratio PDF(x) / max_PDF(x*), that is, the normalized PDF in the range [0,1].

Definition at line 434 of file CPosePDFSOG.cpp.

References mrpt::math::normalPDF().

Here is the call graph for this function:

◆ evaluatePDF()

double CPosePDFSOG::evaluatePDF ( const mrpt::poses::CPose2D x,
bool  sumOverAllPhis = false 
) const

Evaluates the PDF at a given point.

Definition at line 389 of file CPosePDFSOG.cpp.

References mrpt::math::normalPDF(), mrpt::poses::CPoseOrPoint< DERIVEDCLASS, DIM >::x(), and mrpt::poses::CPoseOrPoint< DERIVEDCLASS, DIM >::y().

Here is the call graph for this function:

◆ evaluatePDFInArea()

void CPosePDFSOG::evaluatePDFInArea ( double  x_min,
double  x_max,
double  y_min,
double  y_max,
double  resolutionXY,
double  phi,
mrpt::math::CMatrixDouble outMatrix,
bool  sumOverAllPhis = false 
)

Evaluates the PDF within a rectangular grid (and a fixed orientation) and saves the result in a matrix (each row contains values for a fixed y-coordinate value).

Definition at line 485 of file CPosePDFSOG.cpp.

References ASSERT_, MRPT_END, MRPT_START, and mrpt::math::CMatrixDynamic< T >::setSize().

Here is the call graph for this function:

◆ get() [1/2]

const TGaussianMode& mrpt::poses::CPosePDFSOG::get ( size_t  i) const
inline

Access to individual beacons.

Definition at line 105 of file CPosePDFSOG.h.

References ASSERT_, and m_modes.

◆ get() [2/2]

TGaussianMode& mrpt::poses::CPosePDFSOG::get ( size_t  i)
inline

Access to individual beacons.

Definition at line 111 of file CPosePDFSOG.h.

References ASSERT_, and m_modes.

◆ getAs3DObject() [1/2]

template<class OPENGL_SETOFOBJECTSPTR >
void mrpt::poses::CPosePDF::getAs3DObject ( OPENGL_SETOFOBJECTSPTR &  out_obj) const
inlineinherited

Returns a 3D representation of this PDF (it doesn't clear the current contents of out_obj, but append new OpenGL objects to that list)

Note
Needs the mrpt-opengl library, and using mrpt::opengl::CSetOfObjects::Ptr as template argument.
By default, ellipsoids for the confidence intervals of "q=3" are drawn; for more mathematical details, see CGeneralizedEllipsoidTemplate::setQuantiles()

Definition at line 113 of file CPosePDF.h.

References mrpt::opengl::posePDF2opengl().

Here is the call graph for this function:

◆ getAs3DObject() [2/2]

template<class OPENGL_SETOFOBJECTSPTR >
OPENGL_SETOFOBJECTSPTR mrpt::poses::CPosePDF::getAs3DObject ( ) const
inlineinherited

Returns a 3D representation of this PDF.

Note
Needs the mrpt-opengl library, and using mrpt::opengl::CSetOfObjects::Ptr as template argument.

Definition at line 124 of file CPosePDF.h.

References mrpt::opengl::posePDF2opengl().

Here is the call graph for this function:

◆ getClassName()

static constexpr auto mrpt::poses::CPosePDFSOG::getClassName ( )
inlinestatic

Definition at line 36 of file CPosePDFSOG.h.

◆ getCovariance() [1/3]

void mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::getCovariance ( mrpt::math::CMatrixDouble cov) const
inlineinherited

Returns the estimate of the covariance matrix (STATE_LEN x STATE_LEN covariance matrix)

See also
getMean, getCovarianceAndMean, getInformationMatrix

Definition at line 88 of file CProbabilityDensityFunction.h.

References mrpt::math::CProbabilityDensityFunction< TDATA, STATE_LEN >::getCovarianceDynAndMean().

◆ getCovariance() [2/3]

void mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::getCovariance ( cov_mat_t cov) const
inlineinherited

Returns the estimate of the covariance matrix (STATE_LEN x STATE_LEN covariance matrix)

See also
getMean, getCovarianceAndMean, getInformationMatrix

Definition at line 98 of file CProbabilityDensityFunction.h.

References mrpt::math::CProbabilityDensityFunction< TDATA, STATE_LEN >::getCovarianceAndMean().

◆ getCovariance() [3/3]

cov_mat_t mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::getCovariance ( ) const
inlineinherited

Returns the estimate of the covariance matrix (STATE_LEN x STATE_LEN covariance matrix)

See also
getMean, getInformationMatrix

Definition at line 108 of file CProbabilityDensityFunction.h.

References mrpt::math::cov(), mrpt::math::CProbabilityDensityFunction< TDATA, STATE_LEN >::getCovarianceAndMean(), and mrpt::math::UNINITIALIZED_MATRIX.

◆ getCovarianceAndMean() [1/2]

virtual void mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::getCovarianceAndMean ( cov_mat_t c,
CPose2D &  mean 
) const
inlinefinalvirtualinherited

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Definition at line 54 of file CProbabilityDensityFunction.h.

References mrpt::math::CProbabilityDensityFunction< TDATA, STATE_LEN >::getCovarianceAndMean(), and mrpt::math::mean().

◆ getCovarianceAndMean() [2/2]

std::tuple< CMatrixDouble33, CPose2D > CPosePDFSOG::getCovarianceAndMean ( ) const
overridevirtual

Returns an estimate of the pose covariance matrix (STATE_LENxSTATE_LEN cov matrix) and the mean, both at once.

See also
getMean, getInformationMatrix

Implements mrpt::math::CProbabilityDensityFunction< CPose2D, 3 >.

Definition at line 68 of file CPosePDFSOG.cpp.

References mrpt::math::cov(), mrpt::math::MatrixBase< Scalar, Derived >::matProductOf_AAt(), and mrpt::math::MatrixVectorBase< Scalar, Derived >::setZero().

Here is the call graph for this function:

◆ getCovarianceDynAndMean()

void mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::getCovarianceDynAndMean ( mrpt::math::CMatrixDouble cov,
type_value mean_point 
) const
inlineinherited

Returns an estimate of the pose covariance matrix (STATE_LENxSTATE_LEN cov matrix) and the mean, both at once.

See also
getMean, getInformationMatrix

Definition at line 65 of file CProbabilityDensityFunction.h.

References mrpt::math::cov(), mrpt::math::CProbabilityDensityFunction< TDATA, STATE_LEN >::getCovarianceAndMean(), and mrpt::math::UNINITIALIZED_MATRIX.

◆ getCovarianceEntropy()

double mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::getCovarianceEntropy ( ) const
inlineinherited

◆ getInformationMatrix()

virtual void mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::getInformationMatrix ( inf_mat_t inf) const
inlinevirtualinherited

Returns the information (inverse covariance) matrix (a STATE_LEN x STATE_LEN matrix) Unless reimplemented in derived classes, this method first reads the covariance, then invert it.

See also
getMean, getCovarianceAndMean

Definition at line 130 of file CProbabilityDensityFunction.h.

References mrpt::math::CProbabilityDensityFunction< TDATA, STATE_LEN >::getCovariance(), and mrpt::math::MatrixBase< Scalar, Derived >::inverse_LLt().

◆ getMean() [1/2]

virtual void mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::getMean ( type_value mean_point) const
pure virtualinherited

Returns the mean, or mathematical expectation of the probability density distribution (PDF).

See also
getCovarianceAndMean, getInformationMatrix

◆ getMean() [2/2]

void CPosePDFSOG::getMean ( CPose2D mean_pose) const
override

Definition at line 50 of file CPosePDFSOG.cpp.

References mrpt::poses::SE_average< 2 >::append(), and mrpt::poses::SE_average< 2 >::get_average().

Here is the call graph for this function:

◆ getMeanVal()

type_value mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::getMeanVal ( ) const
inlineinherited

Returns the mean, or mathematical expectation of the probability density distribution (PDF).

See also
getCovariance, getInformationMatrix

Definition at line 77 of file CProbabilityDensityFunction.h.

References mrpt::math::CProbabilityDensityFunction< TDATA, STATE_LEN >::getMean().

◆ getMostLikelyCovarianceAndMean()

void CPosePDFSOG::getMostLikelyCovarianceAndMean ( mrpt::math::CMatrixDouble33 cov,
CPose2D mean_point 
) const

For the most likely Gaussian mode in the SOG, returns the pose covariance matrix (3x3 cov matrix) and the mean.

See also
getMean

Definition at line 645 of file CPosePDFSOG.cpp.

References mrpt::math::CMatrixDynamic< T >::asEigen(), mrpt::containers::begin(), mrpt::math::cov(), mrpt::containers::end(), and mrpt::math::MatrixBase< Scalar, Derived >::setIdentity().

Here is the call graph for this function:

◆ GetRuntimeClass()

virtual const mrpt::rtti::TRuntimeClassId* mrpt::poses::CPosePDFSOG::GetRuntimeClass ( ) const
overridevirtual

Returns information about the class of an object in runtime.

Reimplemented from mrpt::poses::CPosePDF.

◆ GetRuntimeClassIdStatic()

static const mrpt::rtti::TRuntimeClassId& mrpt::poses::CPosePDFSOG::GetRuntimeClassIdStatic ( )
static

◆ getSOGModes()

const CListGaussianModes& mrpt::poses::CPosePDFSOG::getSOGModes ( ) const
inline

Definition at line 69 of file CPosePDFSOG.h.

References m_modes.

◆ inverse()

void CPosePDFSOG::inverse ( CPosePDF o) const
overridevirtual

Returns a new PDF such as: NEW_PDF = (0,0,0) - THIS_PDF.

Implements mrpt::poses::CPosePDF.

Definition at line 355 of file CPosePDFSOG.cpp.

References ASSERT_, CLASS_ID, mrpt::math::cov(), mrpt::poses::CPosePDF::GetRuntimeClass(), mrpt::math::mean(), and out.

Here is the call graph for this function:

◆ is_3D()

static constexpr bool mrpt::poses::CPosePDF::is_3D ( )
inlinestaticinherited

Definition at line 98 of file CPosePDF.h.

References mrpt::poses::CPosePDF::is_3D_val.

◆ is_PDF()

static constexpr bool mrpt::poses::CPosePDF::is_PDF ( )
inlinestaticinherited

Definition at line 103 of file CPosePDF.h.

References mrpt::poses::CPosePDF::is_PDF_val.

◆ isInfType()

virtual bool mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::isInfType ( ) const
inlinevirtualinherited

Returns whether the class instance holds the uncertainty in covariance or information form.

Note
By default this is going to be covariance form. *Inf classes (e.g. CPosePDFGaussianInf) store it in information form.
See also
mrpt::traits::is_inf_type

Reimplemented in mrpt::poses::CPosePDFGaussianInf.

Definition at line 123 of file CProbabilityDensityFunction.h.

◆ jacobiansPoseComposition() [1/2]

void CPosePDF::jacobiansPoseComposition ( const CPose2D x,
const CPose2D u,
mrpt::math::CMatrixDouble33 df_dx,
mrpt::math::CMatrixDouble33 df_du,
const bool  compute_df_dx = true,
const bool  compute_df_du = true 
)
staticinherited

This static method computes the pose composition Jacobians, with these formulas:

df_dx =
[ 1, 0, -sin(phi_x)*x_u-cos(phi_x)*y_u ]
[ 0, 1, cos(phi_x)*x_u-sin(phi_x)*y_u ]
[ 0, 0, 1 ]
df_du =
[ cos(phi_x) , -sin(phi_x) , 0 ]
[ sin(phi_x) , cos(phi_x) , 0 ]
[ 0 , 0 , 1 ]

Definition at line 32 of file CPosePDF.cpp.

References mrpt::poses::CPose2D::phi(), mrpt::math::MatrixBase< Scalar, Derived >::setIdentity(), mrpt::poses::CPoseOrPoint< DERIVEDCLASS, DIM >::x(), and mrpt::poses::CPoseOrPoint< DERIVEDCLASS, DIM >::y().

Referenced by mrpt::poses::CPosePDFGaussian::composePoint(), mrpt::poses::CPosePDF::jacobiansPoseComposition(), mrpt::poses::CPosePDFGaussian::operator+=(), and mrpt::poses::CPosePDFGaussianInf::operator+=().

Here is the call graph for this function:
Here is the caller graph for this function:

◆ jacobiansPoseComposition() [2/2]

void CPosePDF::jacobiansPoseComposition ( const CPosePDFGaussian x,
const CPosePDFGaussian u,
mrpt::math::CMatrixDouble33 df_dx,
mrpt::math::CMatrixDouble33 df_du 
)
staticinherited

This is an overloaded member function, provided for convenience. It differs from the above function only in what argument(s) it accepts.

Definition at line 22 of file CPosePDF.cpp.

References mrpt::poses::CPosePDF::jacobiansPoseComposition().

Here is the call graph for this function:

◆ mergeModes()

void CPosePDFSOG::mergeModes ( double  max_KLd = 0.5,
bool  verbose = false 
)

Merge very close modes so the overall number of modes is reduced while preserving the total distribution.

This method uses the approach described in the paper:

  • "Kullback-Leibler Approach to Gaussian Mixture Reduction" AR Runnalls. IEEE Transactions on Aerospace and Electronic Systems, 2007.
Parameters
max_KLdThe maximum KL-divergence to consider the merge of two nodes (and then stops the process).

Definition at line 516 of file CPosePDFSOG.cpp.

References mrpt::math::CMatrixFixed< T, ROWS, COLS >::asEigen(), mrpt::math::CMatrixDynamic< T >::asEigen(), ASSERT_, mrpt::poses::CPosePDFSOG::TGaussianMode::cov, mrpt::math::cov(), mrpt::math::MatrixBase< Scalar, Derived >::det(), mrpt::poses::CPosePDFSOG::TGaussianMode::log_w, mrpt::math::MatrixBase< Scalar, Derived >::matProductOf_AAt(), mrpt::poses::CPosePDFSOG::TGaussianMode::mean, mrpt::math::mean(), MRPT_END, MRPT_START, mrpt::poses::CPose2D::phi(), verbose, mrpt::math::wrapToPiInPlace(), mrpt::poses::CPoseOrPoint< DERIVEDCLASS, DIM >::x(), and mrpt::poses::CPoseOrPoint< DERIVEDCLASS, DIM >::y().

Here is the call graph for this function:

◆ normalizeWeights()

void CPosePDFSOG::normalizeWeights ( )

Normalize the weights in m_modes such as the maximum log-weight is 0.

Definition at line 468 of file CPosePDFSOG.cpp.

References MRPT_END, and MRPT_START.

◆ operator+=()

void CPosePDFSOG::operator+= ( const mrpt::poses::CPose2D Ap)

Makes: thisPDF = thisPDF + Ap, where "+" is pose composition (both the mean, and the covariance matrix are updated).

Definition at line 379 of file CPosePDFSOG.cpp.

References mrpt::poses::CPose2D::phi().

Here is the call graph for this function:

◆ operator[]() [1/2]

const TGaussianMode& mrpt::poses::CPosePDFSOG::operator[] ( size_t  i) const
inline

Access to individual beacons.

Definition at line 92 of file CPosePDFSOG.h.

References ASSERT_, and m_modes.

◆ operator[]() [2/2]

TGaussianMode& mrpt::poses::CPosePDFSOG::operator[] ( size_t  i)
inline

Access to individual beacons.

Definition at line 98 of file CPosePDFSOG.h.

References ASSERT_, and m_modes.

◆ push_back()

void mrpt::poses::CPosePDFSOG::push_back ( const TGaussianMode m)
inline

Inserts a copy of the given mode into the SOG.

Definition at line 118 of file CPosePDFSOG.h.

References m_modes.

◆ resize()

void CPosePDFSOG::resize ( const size_t  N)

Resize the number of SOG modes.

Definition at line 44 of file CPosePDFSOG.cpp.

Referenced by copyFrom().

Here is the caller graph for this function:

◆ rotateAllCovariances()

void CPosePDFSOG::rotateAllCovariances ( double  ang)

Rotate all the covariance matrixes by replacing them by $ \mathbf{R}~\mathbf{COV}~\mathbf{R}^t $, where $ \mathbf{R} = \left[ \begin{array}{ccc} \cos\alpha & -\sin\alpha & 0 \\ \sin\alpha & \cos\alpha & 0 \\ 0 & 0 & 1 \end{array}\right] $.

Definition at line 233 of file CPosePDFSOG.cpp.

References mrpt::math::multiply_HCHt().

Here is the call graph for this function:

◆ saveToTextFile()

bool CPosePDFSOG::saveToTextFile ( const std::string &  file) const
overridevirtual

Save the density to a text file, with the following format: There is one row per Gaussian "mode", and each row contains 10 elements:

  • w (The weight)
  • x_mean (gaussian mean value)
  • y_mean (gaussian mean value)
  • phi_mean (gaussian mean value)
  • C11 (Covariance elements)
  • C22 (Covariance elements)
  • C33 (Covariance elements)
  • C12 (Covariance elements)
  • C13 (Covariance elements)
  • C23 (Covariance elements)

Implements mrpt::math::CProbabilityDensityFunction< CPose2D, 3 >.

Definition at line 182 of file CPosePDFSOG.cpp.

References mrpt::system::os::fclose(), mrpt::system::os::fopen(), and mrpt::system::os::fprintf().

Here is the call graph for this function:

◆ serializeFrom() [1/2]

void CPosePDFSOG::serializeFrom ( mrpt::serialization::CArchive in,
uint8_t  serial_version 
)
overrideprotectedvirtual

Pure virtual method for reading (deserializing) from an abstract archive.

Users don't call this method directly. Instead, use stream >> object;.

Parameters
inThe input binary stream where the object data must read from.
versionThe version of the object stored in the stream: use this version number in your code to know how to read the incoming data.
Exceptions
std::exceptionOn any I/O error

Implements mrpt::serialization::CSerializable.

Definition at line 118 of file CPosePDFSOG.cpp.

References mrpt::math::CMatrixFixed< T, ROWS, COLS >::cast_double(), mrpt::math::deserializeSymmetricMatrixFrom(), MRPT_THROW_UNKNOWN_SERIALIZATION_VERSION, and resize().

Here is the call graph for this function:

◆ serializeFrom() [2/2]

virtual void mrpt::serialization::CSerializable::serializeFrom ( CSchemeArchiveBase in)
inlineprotectedvirtualinherited

Virtual method for reading (deserializing) from an abstract schema based archive.

Definition at line 74 of file CSerializable.h.

References mrpt::serialization::CSerializable::GetRuntimeClass(), and THROW_EXCEPTION.

Here is the call graph for this function:

◆ serializeGetVersion()

uint8_t CPosePDFSOG::serializeGetVersion ( ) const
overrideprotectedvirtual

Must return the current versioning number of the object.

Start in zero for new classes, and increments each time there is a change in the stored format.

Implements mrpt::serialization::CSerializable.

Definition at line 106 of file CPosePDFSOG.cpp.

◆ serializeTo() [1/2]

void CPosePDFSOG::serializeTo ( mrpt::serialization::CArchive out) const
overrideprotectedvirtual

Pure virtual method for writing (serializing) to an abstract archive.

Users don't call this method directly. Instead, use stream << object;.

Exceptions
std::exceptionOn any I/O error

Implements mrpt::serialization::CSerializable.

Definition at line 107 of file CPosePDFSOG.cpp.

References out, and mrpt::math::serializeSymmetricMatrixTo().

Here is the call graph for this function:

◆ serializeTo() [2/2]

virtual void mrpt::serialization::CSerializable::serializeTo ( CSchemeArchiveBase out) const
inlineprotectedvirtualinherited

Virtual method for writing (serializing) to an abstract schema based archive.

Definition at line 64 of file CSerializable.h.

References mrpt::serialization::CSerializable::GetRuntimeClass(), and THROW_EXCEPTION.

Here is the call graph for this function:

◆ size()

size_t mrpt::poses::CPosePDFSOG::size ( ) const
inline

Return the number of Gaussian modes.

Definition at line 85 of file CPosePDFSOG.h.

References m_modes.

◆ writeToMatlab()

virtual mxArray* mrpt::serialization::CSerializable::writeToMatlab ( ) const
inlinevirtualinherited

Introduces a pure virtual method responsible for writing to a mxArray Matlab object, typically a MATLAB struct whose contents are documented in each derived class.

Returns
A new mxArray (caller is responsible of memory freeing) or nullptr is class does not support conversion to MATLAB.

Definition at line 90 of file CSerializable.h.

Member Data Documentation

◆ className

constexpr const char* mrpt::poses::CPosePDFSOG::className = "mrpt::poses" "::" "CPosePDFSOG"
static

Definition at line 36 of file CPosePDFSOG.h.

◆ m_modes

CListGaussianModes mrpt::poses::CPosePDFSOG::m_modes
protected

The list of SOG modes.

Definition at line 77 of file CPosePDFSOG.h.

Referenced by begin(), empty(), end(), erase(), get(), getSOGModes(), operator[](), push_back(), and size().

◆ runtimeClassId

const mrpt::rtti::TRuntimeClassId mrpt::poses::CPosePDFSOG::runtimeClassId
staticprotected

Definition at line 36 of file CPosePDFSOG.h.

◆ state_length

constexpr size_t mrpt::math::CProbabilityDensityFunction< CPose2D , STATE_LEN >::state_length
staticinherited

The length of the variable, for example, 3 for a 3D point, 6 for a 3D pose (x y z yaw pitch roll).

Definition at line 32 of file CProbabilityDensityFunction.h.




Page generated by Doxygen 1.8.14 for MRPT 2.0.2 Git: 9b4fd2465 Mon May 4 16:59:08 2020 +0200 at lun may 4 17:26:07 CEST 2020