CBeacon.h

Go to the documentation of this file.

/* +------------------------------------------------------------------------+
   |                     Mobile Robot Programming Toolkit (MRPT)            |
   |                          https://www.mrpt.org/                         |
   |                                                                        |
   | Copyright (c) 2005-2020, Individual contributors, see AUTHORS file     |
   | See: https://www.mrpt.org/Authors - All rights reserved.               |
   | Released under BSD License. See: https://www.mrpt.org/License          |
   +------------------------------------------------------------------------+ */
#pragma once

#include <mrpt/math/CMatrixF.h>
#include <mrpt/obs/CObservation.h>  // INVALID_BEACON_ID
#include <mrpt/poses/CPoint3D.h>
#include <mrpt/poses/CPointPDFGaussian.h>
#include <mrpt/poses/CPointPDFParticles.h>
#include <mrpt/poses/CPointPDFSOG.h>
#include <mrpt/serialization/CSerializable.h>

#include <mrpt/opengl/CSetOfObjects.h>

namespace mrpt::maps
{
class CBeaconMap;

/** The class for storing individual "beacon landmarks" under a variety of 3D
 * position PDF distributions.
 *  This class is used for storage within the class CBeaconMap.
 *  The class implements the same methods than the interface "CPointPDF", and
 * invoking them actually becomes
 *   a mapping into the methods of the current PDF representation of the
 * beacon, selectable by means of "m_typePDF"
 * \sa CBeaconMap, CPointPDFSOG
 * \ingroup mrpt_maps_grp
 */
class CBeacon : public mrpt::poses::CPointPDF
{
    DEFINE_SERIALIZABLE(CBeacon, mrpt::maps)

   public:
    /** The type for the IDs of landmarks.
     */
    using TBeaconID = int64_t;

    /** See m_typePDF
     */
    enum TTypePDF
    {
        pdfMonteCarlo = 0,
        pdfGauss,
        pdfSOG
    };

    /** Which one of the different 3D point PDF is currently used in this
     * object: montecarlo, gaussian, or a sum of gaussians.
     * \sa m_location
     */
    TTypePDF m_typePDF{pdfGauss};

    /** The individual PDF, if m_typePDF=pdfMonteCarlo (publicly accesible for
     * ease of use, but the CPointPDF interface is also implemented in CBeacon).
     */
    mrpt::poses::CPointPDFParticles m_locationMC{1};
    /** The individual PDF, if m_typePDF=pdfGauss (publicly accesible for ease
     * of use, but the CPointPDF interface is also implemented in CBeacon). */
    mrpt::poses::CPointPDFGaussian m_locationGauss;
    /** The individual PDF, if m_typePDF=pdfSOG (publicly accesible for ease of
     * use, but the CPointPDF interface is also implemented in CBeacon). */
    mrpt::poses::CPointPDFSOG m_locationSOG{1};

    /** An ID for the landmark (see details next...)
     *  This ID was introduced in the version 3 of this class (21/NOV/2006),
     *and its aim is
     *  to provide a way for easily establishing correspondences between
     *landmarks detected
     *  in sequential image frames. Thus, the management of this field should
     *be:
     *      - In 'servers' (classes/modules/... that detect landmarks from
     *images):
     *A different ID must be assigned to every landmark (e.g. a sequential
     *counter), BUT only in the case of being sure of the correspondence of one
     *landmark with another one in the past (e.g. tracking).
     *      - In 'clients': This field can be ignored, but if it is used, the
     *advantage is solving the correspondence between landmarks detected in
     *consequentive instants of time: Two landmarks with the same ID
     *<b>correspond</b> to the same physical feature, BUT it should not be
     *expected the inverse to be always true.
     *
     * Note that this field is never fill out automatically, it must be set by
     *the programmer if used.
     */
    TBeaconID m_ID{INVALID_BEACON_ID};

    void getMean(mrpt::poses::CPoint3D& mean_point) const override;

    std::tuple<cov_mat_t, type_value> getCovarianceAndMean() const override;

    /** Copy operator, translating if necesary (for example, between particles
     * and gaussian representations) */
    void copyFrom(const mrpt::poses::CPointPDF& o) override;

    /** Save PDF's particles to a text file. See derived classes for more
     * information about the format of generated files */
    bool saveToTextFile(const std::string& file) const override;

    /** 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.
     */
    void changeCoordinatesReference(
        const mrpt::poses::CPose3D& newReferenceBase) override;

    /** Saves a 3D representation of the beacon into a given OpenGL scene  */
    void getAs3DObject(mrpt::opengl::CSetOfObjects::Ptr& outObj) const;

    /** Gets a set of MATLAB commands which draw the current state of the
     * beacon: */
    void getAsMatlabDrawCommands(std::vector<std::string>& out_Str) const;

    /** Draw a sample from the pdf. */
    void drawSingleSample(mrpt::poses::CPoint3D& outSample) const override;

    /** Bayesian fusion of two point distributions (product of two
     * distributions->new distribution), then save the result in this object
     * (WARNING: See implementing classes to see classes that can and cannot be
     * mixtured!)
     * \param p1 The first distribution to fuse
     * \param p2 The second distribution to fuse
     * \param minMahalanobisDistToDrop If set to different of 0, the result of
     * very separate Gaussian modes (that will result in negligible components)
     * in SOGs will be dropped to reduce the number of modes in the output.
     */
    void bayesianFusion(
        const CPointPDF& p1, const CPointPDF& p2,
        const double minMahalanobisDistToDrop = 0) override;

    /** Compute the observation model p(z_t|x_t) for a given observation (range
     * value), and return it as an approximate SOG.
     *  Note that if the beacon is a SOG itself, the number of gaussian modes
     * will be square.
     *  As a speed-up, if a "center point"+"maxDistanceFromCenter" is supplied
     * (maxDistanceFromCenter!=0), those modes farther than this sphere will be
     * discarded.
     *  Parameters such as the stdSigma of the sensor are gathered from
     * "myBeaconMap"
     *  The result is one "ring" for each Gaussian mode that represent the
     * beacon position in this object.
     *  The position of the sensor on the robot is used to shift the resulting
     * densities such as they represent the position of the robot, not the
     * sensor.
     *  \sa CBeaconMap::insertionOptions, generateRingSOG
     */
    void generateObservationModelDistribution(
        float sensedRange, mrpt::poses::CPointPDFSOG& outPDF,
        const CBeaconMap* myBeaconMap,
        const mrpt::poses::CPoint3D& sensorPntOnRobot,
        const mrpt::poses::CPoint3D& centerPoint =
            mrpt::poses::CPoint3D(0, 0, 0),
        float maxDistanceFromCenter = 0) const;

    /** This static method returns a SOG with ring-shape (or as a 3D sphere)
     * that can be used to initialize a beacon if observed the first time.
     *  sensorPnt is the center of the ring/sphere, i.e. the absolute position
     * of the range sensor.
     *  If clearPreviousContentsOutPDF=false, the SOG modes will be added to
     * the current contents of outPDF
     *  If the 3x3 matrix covarianceCompositionToAdd is provided, it will be
     * add to every Gaussian (to model the composition of uncertainty).
     * \sa generateObservationModelDistribution
     */
    static void generateRingSOG(
        float sensedRange, mrpt::poses::CPointPDFSOG& outPDF,
        const CBeaconMap* myBeaconMap, const mrpt::poses::CPoint3D& sensorPnt,
        const mrpt::math::CMatrixDouble33* covarianceCompositionToAdd = nullptr,
        bool clearPreviousContentsOutPDF = true,
        const mrpt::poses::CPoint3D& centerPoint =
            mrpt::poses::CPoint3D(0, 0, 0),
        float maxDistanceFromCenter = 0);

};  // End of class definition

}  // namespace mrpt::maps