Class CMetricMapBuilderICP

Nested Relationships

Nested Types

Inheritance Relationships

Base Type

Class Documentation

class CMetricMapBuilderICP : public mrpt::slam::CMetricMapBuilder

A class for very simple 2D SLAM based on ICP. This is a non-probabilistic pose tracking algorithm. Map are stored as in files as binary dumps of “mrpt::maps::CSimpleMap” objects. The methods are thread-safe.

Public Functions

CMetricMapBuilderICP()

Default constructor - Upon construction, you can set the parameters in ICP_options, then call “initialize”.

~CMetricMapBuilderICP() override

Destructor:

virtual void initialize(const mrpt::maps::CSimpleMap &initialMap = mrpt::maps::CSimpleMap(), const mrpt::poses::CPosePDF *x0 = nullptr) override

Initialize the method, starting with a known location PDF “x0”(if supplied, set to nullptr to left unmodified) and a given fixed, past map. This method MUST be called if using the default constructor, after loading the configuration into ICP_options. In particular, TConfigParams::mapInitializers

virtual mrpt::poses::CPose3DPDF::Ptr getCurrentPoseEstimation() const override

Returns a copy of the current best pose estimation as a pose PDF.

void setCurrentMapFile(const char *mapFile)

Sets the “current map file”, thus that map will be loaded if it exists or a new one will be created if it does not, and the updated map will be save to that file when destroying the object.

virtual void processActionObservation(mrpt::obs::CActionCollection &action, mrpt::obs::CSensoryFrame &in_SF) override

Appends a new action and observations to update this map: See the description of the class at the top of this page to see a more complete description.

See also

processObservation

Parameters:

  • action – The estimation of the incremental pose change in the robot pose.

  • in_SF – The set of observations that robot senses at the new pose. See params in CMetricMapBuilder::options and CMetricMapBuilderICP::ICP_options

void processObservation(const mrpt::obs::CObservation::Ptr &obs)

The main method of this class: Process one odometry or sensor observation. The new entry point of the algorithm (the old one was processActionObservation, which now is a wrapper to this method). See params in CMetricMapBuilder::options and CMetricMapBuilderICP::ICP_options

virtual void getCurrentlyBuiltMap(mrpt::maps::CSimpleMap &out_map) const override

Fills “out_map” with the set of “poses”-“sensory-frames”, thus the so far built map

void getCurrentMapPoints(std::vector<float> &x, std::vector<float> &y)

Returns the 2D points of current local map

virtual const mrpt::maps::CMultiMetricMap &getCurrentlyBuiltMetricMap() const override

Returns the map built so far. NOTE that for efficiency a pointer to the internal object is passed, DO NOT delete nor modify the object in any way, if desired, make a copy of ir with “clone()”.

virtual unsigned int getCurrentlyBuiltMapSize() override

Returns just how many sensory-frames are stored in the currently build map

virtual void saveCurrentEstimationToImage(const std::string &file, bool formatEMF_BMP = true) override

A useful method for debugging: the current map (and/or poses) estimation is dumped to an image file.

Parameters:

  • file – The output file name

  • formatEMF_BMP – Output format = true:EMF, false:BMP

Public Members

TConfigParams ICP_options

Options for the ICP-SLAM application

See also

ICP_params

CICP::TConfigParams ICP_params

Options for the ICP algorithm itself

See also

ICP_options

struct TConfigParams : public mrpt::config::CLoadableOptions

Algorithm configuration params

Public Functions

TConfigParams(mrpt::system::VerbosityLevel &parent_verbosity_level)

Initializer

TConfigParams &operator=(const TConfigParams &other)
inline TConfigParams(const TConfigParams &other)
void loadFromConfigFile(const mrpt::config::CConfigFileBase &source, const std::string &section) override
void dumpToTextStream(std::ostream &out) const override

Public Members

bool matchAgainstTheGrid

(default:false) Match against the occupancy grid or the points map? The former is quicker but less precise.

double insertionLinDistance

Minimum robot linear (m) displacement for a new observation to be inserted in the map.

double insertionAngDistance

Minimum robot angular (rad, deg when loaded from the .ini) displacement for a new observation to be inserted in the map.

double localizationLinDistance

Minimum robot linear (m) displacement for a new observation to be used to do ICP-based localization (otherwise, dead-reckon with odometry).

double localizationAngDistance

Minimum robot angular (rad, deg when loaded from the .ini) displacement for a new observation to be used to do ICP-based localization (otherwise, dead-reckon with odometry).

double minICPgoodnessToAccept

Minimum ICP goodness (0,1) to accept the resulting corrected position (default: 0.40)

mrpt::system::VerbosityLevel &verbosity_level
mrpt::maps::TSetOfMetricMapInitializers mapInitializers

What maps to create (at least one points map and/or a grid map are needed). For the expected format in the .ini file when loaded with loadFromConfigFile(), see documentation of TSetOfMetricMapInitializers.