struct mp2p_icp_filters::Generator::Parameters
Overview
#include <Generator.h> struct Parameters { // fields std::string target_layer = mp2p_icp::metric_map_t::PT_LAYER_RAW; std::string metric_map_definition_ini_file; mrpt::containers::yaml metric_map_definition; std::string process_class_names_regex = ".*"; std::string process_sensor_labels_regex = ".*"; bool throw_on_unhandled_observation_class = false; // methods void load_from_yaml( const mrpt::containers::yaml& c, Generator& parent ); };
Fields
std::string target_layer = mp2p_icp::metric_map_t::PT_LAYER_RAW
The map layer name where the observation will be loaded into. Default: “raw”
std::string metric_map_definition_ini_file
If empty (default), a point cloud layer will be generated of type mrpt::maps::CSimplePointCloud, and the observation will be inserted via the virtual method mrpt::maps::CMetricMap::insertObservation()
Alternatively, if this variable or metric_map_definition
are set, a custom metric map class can be defined via a mrpt::maps::CMultiMetricMap initializer list. If the CMultiMetricMap defines multiple metric maps, the first one only will be taken to generate the new layer.
This variable may be set to an external .ini
file in the format expected by mrpt::maps::TSetOfMetricMapInitializers. The recomended alternative is to directly define the map setting in the YAML file.
Refer to example pipelines.
mrpt::containers::yaml metric_map_definition
Read metric_map_definition_ini_file. Defining a custom metric map via this YAML structure is preferred, since parameterizable variables are supported in this way, while external .ini
files must contain fixed static values only.
std::string process_class_names_regex = ".*"
Sensor observation class names to process. Default = “.*” (any). Example: use “mrpt::obs::CObservation2DRangeScan” if you only want 2D lidar scans to be processed.
std::string process_sensor_labels_regex = ".*"
Sensor labels to process. Default = “.*” (any). Examples: “LIDAR”, “FRONT_KINECT”, “LIDAR_.*”