autoware::localization::ndt Namespace Reference

Namespaces
	traits
	transform_adapters

Classes
class	DynamicNDTMap
class	DynamicNDTVoxel
struct	geocentric_pose_t
struct	geodetic_pose_t
class	NDTGrid
	A voxel grid implementation for normal distribution transform. More...
class	NDTLocalizerBase
class	NDTLocalizerConfigBase
class	NDTScanBase
class	P2DNDTLocalizer
class	P2DNDTLocalizerConfig
	config class for p2d ndt localizer More...
class	P2DNDTObjective
class	P2DNDTOptimizationConfig
	Config class for p2d optimziation problem. More...
class	P2DNDTScan
struct	PointWithCovariances
class	StaticNDTMap
class	StaticNDTVoxel
class	VoxelView< DynamicNDTVoxel >
class	VoxelView< StaticNDTVoxel >
	VoxelViewBase implementation for StaticNDTVoxel. It's just a pure wrapper. More...
class	VoxelViewBase

Typedefs
using	Real = float64_t
using	CloudT = sensor_msgs::msg::PointCloud2
template<typename MapT >
using	P2DNDTOptimizationProblem = common::optimization::UnconstrainedOptimizationProblem< P2DNDTObjective< MapT >, EigenPose< Real >, 6U >
using	PointWithCovariancesFieldGenerators = std::tuple< point_cloud_msg_wrapper::field_x_generator, point_cloud_msg_wrapper::field_y_generator, point_cloud_msg_wrapper::field_z_generator, field_icov_xx_generator, field_icov_xy_generator, field_icov_xz_generator, field_icov_yy_generator, field_icov_yz_generator, field_icov_zz_generator >
using	NdtMapCloudModifier = point_cloud_msg_wrapper::PointCloud2Modifier< PointWithCovariances, PointWithCovariancesFieldGenerators >
using	NdtMapCloudView = point_cloud_msg_wrapper::PointCloud2View< PointWithCovariances, PointWithCovariancesFieldGenerators >
template<typename T >
using	EigenPose = Eigen::Matrix< T, 6U, 1U >
template<typename T >
using	EigenTransform = Eigen::Transform< T, 3, Eigen::Affine, Eigen::ColMajor >
using	RosTransform = geometry_msgs::msg::Transform
using	RosPose = geometry_msgs::msg::Pose
using	StaticView = VoxelView< StaticNDTVoxel >
using	DynamicView = VoxelView< DynamicNDTVoxel >

Enumerations
enum	Requires
enum	Invertibility { Invertibility::INVERTIBLE, Invertibility::NOT_INVERTIBLE, Invertibility::UNKNOWN }

Functions
void NDT_PUBLIC	read_from_yaml (const std::string &yaml_file_name, geodetic_pose_t *geo_pose)
void NDT_PUBLIC	read_from_pcd (const std::string &file_name, sensor_msgs::msg::PointCloud2 *msg)
geocentric_pose_t NDT_PUBLIC	load_map (const std::string &yaml_file_name, const std::string &pcl_file_name, sensor_msgs::msg::PointCloud2 &pc_out)
	Read the pcd file with filename into a PointCloud2 message, transform it into an NDT representation and then serialize the ndt representation back into a PointCloud2 message that can be published. More...
template<typename ScalarT >
bool8_t	is_valid_probability (ScalarT p)
	LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER (icov_xx)
	LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER (icov_xy)
	LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER (icov_xz)
	LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER (icov_yy)
	LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER (icov_yz)
	LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER (icov_zz)
template<typename Derived >
bool	try_stabilize_covariance (Eigen::MatrixBase< Derived > &covariance, typename Derived::PlainMatrix::Scalar scaling_factor=0.1)

Variables
template<typename VoxelT >
class NDT_PUBLIC	VoxelView

Typedef Documentation

CloudT

using autoware::localization::ndt::CloudT = typedef sensor_msgs::msg::PointCloud2

DynamicView

using autoware::localization::ndt::DynamicView = typedef VoxelView<DynamicNDTVoxel>

EigenPose

template<typename T >

using autoware::localization::ndt::EigenPose = typedef Eigen::Matrix<T, 6U, 1U>

EigenTransform

template<typename T >

using autoware::localization::ndt::EigenTransform = typedef Eigen::Transform<T, 3, Eigen::Affine, Eigen::ColMajor>

NdtMapCloudModifier

using autoware::localization::ndt::NdtMapCloudModifier = typedef point_cloud_msg_wrapper::PointCloud2Modifier<PointWithCovariances, PointWithCovariancesFieldGenerators>

NdtMapCloudView

using autoware::localization::ndt::NdtMapCloudView = typedef point_cloud_msg_wrapper::PointCloud2View<PointWithCovariances, PointWithCovariancesFieldGenerators>

P2DNDTOptimizationProblem

template<typename MapT >

using autoware::localization::ndt::P2DNDTOptimizationProblem = typedef common::optimization::UnconstrainedOptimizationProblem<P2DNDTObjective<MapT>, EigenPose<Real>, 6U>

PointWithCovariancesFieldGenerators

using autoware::localization::ndt::PointWithCovariancesFieldGenerators = typedef std::tuple< point_cloud_msg_wrapper::field_x_generator, point_cloud_msg_wrapper::field_y_generator, point_cloud_msg_wrapper::field_z_generator, field_icov_xx_generator, field_icov_xy_generator, field_icov_xz_generator, field_icov_yy_generator, field_icov_yz_generator, field_icov_zz_generator>

Real

using autoware::localization::ndt::Real = typedef float64_t

RosPose

using autoware::localization::ndt::RosPose = typedef geometry_msgs::msg::Pose

RosTransform

using autoware::localization::ndt::RosTransform = typedef geometry_msgs::msg::Transform

StaticView

using autoware::localization::ndt::StaticView = typedef VoxelView<StaticNDTVoxel>

Enumeration Type Documentation

Invertibility

enum autoware::localization::ndt::Invertibility

strong

Enumerator
INVERTIBLE
NOT_INVERTIBLE
UNKNOWN

Requires

enum autoware::localization::ndt::Requires

strong

Function Documentation

is_valid_probability()

template<typename ScalarT >

bool8_t autoware::localization::ndt::is_valid_probability

(

ScalarT

p

)

LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER() [1/6]

autoware::localization::ndt::LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER

(

icov_xx

)

LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER() [2/6]

autoware::localization::ndt::LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER

(

icov_xy

)

LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER() [3/6]

autoware::localization::ndt::LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER

(

icov_xz

)

LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER() [4/6]

autoware::localization::ndt::LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER

(

icov_yy

)

LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER() [5/6]

autoware::localization::ndt::LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER

(

icov_yz

)

LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER() [6/6]

autoware::localization::ndt::LIDAR_UTILS__DEFINE_FIELD_GENERATOR_FOR_MEMBER

(

icov_zz

)

load_map()

geocentric_pose_t autoware::localization::ndt::load_map	(	const std::string &	yaml_file_name,
		const std::string &	pcl_file_name,
		sensor_msgs::msg::PointCloud2 &	pc_out
	)

Read the pcd file with filename into a PointCloud2 message, transform it into an NDT representation and then serialize the ndt representation back into a PointCloud2 message that can be published.

Parameters

yaml_file_name	File name of the yaml file.
pcl_file_name	File name of the pcd file.
pc_out	Reference to the pointcloud message that will be populated from the data on the disk.

Returns

The geocentric position.

read_from_pcd()

void autoware::localization::ndt::read_from_pcd	(	const std::string &	file_name,
		sensor_msgs::msg::PointCloud2 *	msg
	)

Read the pcd file into a PointCloud2 message. Throws if the file cannot be read. The returned point cloud has float fields x, y, z, intensity, all of type float32_t, with the default intensity value being 0.0.

Parameters

[in]	file_name	Name of the pcd file.
[out]	msg	Pointer to PointCloud2 message

read_from_yaml()

void autoware::localization::ndt::read_from_yaml	(	const std::string &	yaml_file_name,
		geodetic_pose_t *	geo_pose
	)

Read the map info from a yaml file. Throws if the file cannot be read.

Parameters

[in]	yaml_file_name	Name of the ymal file.
[out]	geo_pose	Geodetic pose describing map orgin

try_stabilize_covariance()

template<typename Derived >

bool autoware::localization::ndt::try_stabilize_covariance	(	Eigen::MatrixBase< Derived > &	covariance,
		typename Derived::PlainMatrix::Scalar	scaling_factor = 0.1
	)

This function will check if the covariance is valid based on its eigenvalues. If the covariance is valid, eigen values smaller than a fraction of the biggest eigen value will be capped to the threshold. Covariance then will be reconstructed from the modified set of eigen values and vectors. This should result in increased numerical stability as stated in [Magnusson 2009]. If the covariance is invalid, it will not be updated and false will be returned.

Template Parameters

Derived

Deduced Eigen Matrix type.

Parameters

covariance	[in, out] Covariance matrix to get stabilized.
scaling_factor	[in] The ratio between the max. eigen value and the minimum allowed eigenvalue. Default value is 0.01 as suggested in [Magnusson 2009], page 60.

Returns

True if the covariance matrix is valid.

Variable Documentation

VoxelView

template<typename VoxelT >

class NDT_PUBLIC autoware::localization::ndt::VoxelView