Detailed Description

A class used to store a 3D pose (a 3D translation + a rotation in 3D).

The 6D transformation in SE(3) stored in this class is kept in two separate containers: a 3-array for the translation, and a 3x3 rotation matrix.

The 6D pose is parameterized as a 6-vector: [x y z yaw pitch roll]. Note however, that the yaw/pitch/roll angles are only computed (on-demand and transparently) when the user requests them. Normally, rotations are handled via the 3x3 rotation matrix only.

For a complete descriptionan of Points/Poses, see mrpt::poses::CPoseOrPoint, or refer to the 2D/3D Geometry tutorial in the wiki.

To change the individual components of the pose, use CPose3D::setFromValues. This class assures that the internal 3x3 rotation matrix is always up-to-date with the "yaw pitch roll" members.

Rotations in 3D can be also represented by quaternions. See mrpt::math::CQuaternion, and method CPose3D::getAsQuaternion.

This class and CPose3DQuat are very similar, and they can be converted to the each other automatically via transformation constructors.

There are Lie algebra methods: exp and ln (see the methods for documentation).

See also:: CPoseOrPoint,CPoint3D, mrpt::math::CQuaternion

#include <mrpt/poses/CPose3D.h>

Inheritance diagram for mrpt::poses::CPose3D:

[legend]

List of all members.

Public Types
enum	{ is_3D_val = 1 }
enum	{ rotation_dimensions = 3 }
enum	{ is_PDF_val = 0 }
typedef CPose3D	type_value
	Used to emulate CPosePDF types, for example, in mrpt::graphs::CNetworkOfPoses.
Public Member Functions
const type_value &	getPoseMean () const
type_value &	getPoseMean ()
mrpt::utils::CObjectPtr	duplicateGetSmartPtr () const
	Returns a copy of the object, indepently of its class, as a smart pointer (the newly created object will exist as long as any copy of this smart pointer).
CObject *	clone () const
	Cloning interface for smart pointers.
Constructors
	CPose3D ()
	Default constructor, with all the coordinates set to zero.
	CPose3D (const double x, const double y, const double z, const double yaw=0, const double pitch=0, const double roll=0)
	Constructor with initilization of the pose; (remember that angles are always given in radians!)
	CPose3D (const math::CMatrixDouble &m)
	Constructor from a 4x4 homogeneous matrix - the passed matrix can be actually of any size larger than or equal 3x4, since only those first values are used (the last row of a homogeneous 4x4 matrix are always fixed).
	CPose3D (const math::CMatrixDouble44 &m)
	Constructor from a 4x4 homogeneous matrix:
template<class MATRIX33 , class VECTOR3 >
	CPose3D (const MATRIX33 &rot, const VECTOR3 &xyz)
	Constructor from a 3x3 rotation matrix and a the translation given as a 3-vector, a 3-array, a CPoint3D or a TPoint3D.
	CPose3D (const CMatrixDouble33 &rot, const CArrayDouble< 3 > &xyz)
	CPose3D (const CPose2D &)
	Constructor from a CPose2D object.
	CPose3D (const CPoint3D &)
	Constructor from a CPoint3D object.
	CPose3D (const mrpt::math::TPose3D &)
	Constructor from lightweight object.
	CPose3D (const mrpt::math::CQuaternionDouble &q, const double x, const double y, const double z)
	Constructor from a quaternion (which only represents the 3D rotation part) and a 3D displacement.
	CPose3D (const CPose3DQuat &)
	Constructor from a CPose3DQuat.
	CPose3D (TConstructorFlags_Poses constructor_dummy_param)
	Fast constructor that leaves all the data uninitialized - call with UNINITIALIZED_POSE as argument.
	CPose3D (const CArrayDouble< 12 > &vec12)
	Constructor from an array with these 12 elements: [r11 r21 r31 r12 r22 r32 r13 r23 r33 tx ty tz] where r{ij} are the entries of the 3x3 rotation matrix and t{x,y,z} are the 3D translation of the pose.
Access 3x3 rotation and 4x4 homogeneous matrices
void	getHomogeneousMatrix (CMatrixDouble44 &out_HM) const
	Returns the corresponding 4x4 homogeneous transformation matrix for the point(translation) or pose (translation+orientation).
CMatrixDouble44	getHomogeneousMatrixVal () const
void	getRotationMatrix (mrpt::math::CMatrixDouble33 &ROT) const
	Get the 3x3 rotation matrix.
const mrpt::math::CMatrixDouble33 &	getRotationMatrix () const
Pose-pose and pose-point compositions and operators
CPose3D	operator+ (const CPose3D &b) const
	The operator $a \oplus b$ is the pose compounding operator.
CPoint3D	operator+ (const CPoint3D &b) const
	The operator $a \oplus b$ is the pose compounding operator.
CPoint3D	operator+ (const CPoint2D &b) const
	The operator $a \oplus b$ is the pose compounding operator.
void	sphericalCoordinates (const TPoint3D &point, double &out_range, double &out_yaw, double &out_pitch) const
	Computes the spherical coordinates of a 3D point as seen from the 6D pose specified by this object.
void	composePoint (double lx, double ly, double lz, double &gx, double &gy, double &gz, mrpt::math::CMatrixFixedNumeric< double, 3, 3 > out_jacobian_df_dpoint=NULL, mrpt::math::CMatrixFixedNumeric< double, 3, 6 > out_jacobian_df_dpose=NULL, mrpt::math::CMatrixFixedNumeric< double, 3, 6 > *out_jacobian_df_dse3=NULL, bool use_small_rot_approx=false) const
	An alternative, slightly more efficient way of doing $G = P \oplus L$ with G and L being 3D points and P this 6D pose.
void	composePoint (const TPoint3D local_point, TPoint3D &global_point) const
	An alternative, slightly more efficient way of doing $G = P \oplus L$ with G and L being 3D points and P this 6D pose.
void	composePoint (double lx, double ly, double lz, float &gx, float &gy, float &gz) const
	An alternative, slightly more efficient way of doing $G = P \oplus L$ with G and L being 3D points and P this 6D pose.
void	inverseComposePoint (const double gx, const double gy, const double gz, double &lx, double &ly, double &lz, mrpt::math::CMatrixFixedNumeric< double, 3, 3 > out_jacobian_df_dpoint=NULL, mrpt::math::CMatrixFixedNumeric< double, 3, 6 > out_jacobian_df_dpose=NULL, mrpt::math::CMatrixFixedNumeric< double, 3, 6 > *out_jacobian_df_dse3=NULL) const
	Computes the 3D point L such as $L = G \ominus this$ .
void	composeFrom (const CPose3D &A, const CPose3D &B)
	Makes "this = A (+) B"; this method is slightly more efficient than "this= A + B;" since it avoids the temporary object.
CPose3D &	operator+= (const CPose3D &b)
	Make $this = this \oplus b$ (b can be "this" without problems)
void	inverseComposeFrom (const CPose3D &A, const CPose3D &B)
	Makes $this = A \ominus B$ this method is slightly more efficient than "this= A - B;" since it avoids the temporary object.
CPose3D	operator- (const CPose3D &b) const
	Compute $RET = this \oplus b$ .
void	inverse ()
	Convert this pose into its inverse, saving the result in itself.
void	changeCoordinatesReference (const CPose3D &p)
	makes: this = p (+) this
Access and modify contents
void	addComponents (const CPose3D &p)
	Scalar sum of all 6 components: This is diferent from poses composition, which is implemented as "+" operators.
void	normalizeAngles ()
	Rebuild the internal matrix & update the yaw/pitch/roll angles within the ]-PI,PI] range (Must be called after using addComponents)
void	operator*= (const double s)
	Scalar multiplication of x,y,z,yaw,pitch & roll (angles will be wrapped to the ]-pi,pi] interval).
void	setFromValues (const double x0, const double y0, const double z0, const double yaw=0, const double pitch=0, const double roll=0)
	Set the pose from a 3D position (meters) and yaw/pitch/roll angles (radians) - This method recomputes the internal rotation matrix.
template<typename VECTORLIKE >
void	setFromXYZQ (const VECTORLIKE &v, const size_t index_offset=0)
	Set the pose from a 3D position (meters) and a quaternion, stored as [x y z qr qx qy qz] in a 7-element vector.
void	setYawPitchRoll (const double yaw_, const double pitch_, const double roll_)
	Set the 3 angles of the 3D pose (in radians) - This method recomputes the internal rotation coordinates matrix.
template<class ARRAYORVECTOR >
void	setFrom12Vector (const ARRAYORVECTOR &vec12)
	Set pose from an array with these 12 elements: [r11 r21 r31 r12 r22 r32 r13 r23 r33 tx ty tz] where r{ij} are the entries of the 3x3 rotation matrix and t{x,y,z} are the 3D translation of the pose.
template<class ARRAYORVECTOR >
void	getAs12Vector (ARRAYORVECTOR &vec12) const
	Get the pose representation as an array with these 12 elements: [r11 r21 r31 r12 r22 r32 r13 r23 r33 tx ty tz] where r{ij} are the entries of the 3x3 rotation matrix and t{x,y,z} are the 3D translation of the pose.
void	getYawPitchRoll (double &yaw, double &pitch, double &roll) const
	Returns the three angles (yaw, pitch, roll), in radians, from the rotation matrix.
double	yaw () const
	Get the YAW angle (in radians)
double	pitch () const
	Get the PITCH angle (in radians)
double	roll () const
	Get the ROLL angle (in radians)
void	getAsVector (vector_double &v) const
	Returns a 1x6 vector with [x y z yaw pitch roll].
void	getAsQuaternion (mrpt::math::CQuaternionDouble &q, mrpt::math::CMatrixFixedNumeric< double, 4, 3 > *out_dq_dr=NULL) const
	Returns the quaternion associated to the rotation of this object (NOTE: XYZ translation is ignored) $\mathbf{q} = \left( \begin{array}{c} \cos (\phi /2) \cos (\theta /2) \cos (\psi /2) + \sin (\phi /2) \sin (\theta /2) \sin (\psi /2) \\ \sin (\phi /2) \cos (\theta /2) \cos (\psi /2) - \cos (\phi /2) \sin (\theta /2) \sin (\psi /2) \\ \cos (\phi /2) \sin (\theta /2) \cos (\psi /2) + \sin (\phi /2) \cos (\theta /2) \sin (\psi /2) \\ \cos (\phi /2) \cos (\theta /2) \sin (\psi /2) - \sin (\phi /2) \sin (\theta /2) \cos (\psi /2) \\ \end{array}\right)$ With : $\phi = roll$ , $\theta = pitch$ and $\psi = yaw$ .
const double &	operator[] (unsigned int i) const
void	asString (std::string &s) const
	Returns a human-readable textual representation of the object (eg: "[x y z yaw pitch roll]", angles in degrees.)
std::string	asString () const
void	fromString (const std::string &s)
	Set the current object value from a string generated by 'asString' (eg: "[x y z yaw pitch roll]", angles in deg.
bool	isHorizontal (const double tolerance=0) const
	Return true if the 6D pose represents a Z axis almost exactly vertical (upwards or downwards), with a given tolerance (if set to 0 exact horizontality is tested).
double	distanceEuclidean6D (const CPose3D &o) const
	The euclidean distance between two poses taken as two 6-length vectors (angles in radians).
Static Public Member Functions
static bool	is_3D ()
static bool	is_PDF ()
Public Attributes
CArrayDouble< 3 >	m_coords
	The translation vector [x,y,z].
CMatrixDouble33	m_ROT
	The 3x3 rotation matrix.
Static Public Attributes
static const mrpt::utils::TRuntimeClassId	classCObject
Protected Member Functions
void	rebuildRotationMatrix ()
	Rebuild the homog matrix from the angles.
void	updateYawPitchRoll () const
	Updates Yaw/pitch/roll members from the m_ROT.
virtual void	writeToStream (mrpt::utils::CStream &out, int *getVersion) const =0
	Introduces a pure virtual method responsible for writing to a CStream.
virtual void	readFromStream (mrpt::utils::CStream &in, int version)=0
	Introduces a pure virtual method responsible for loading from a CStream This can not be used directly be users, instead use "stream >> object;" for reading it from a stream or "stream >> object_ptr;" if the class is unknown apriori.
Protected Attributes
bool	m_ypr_uptodate
	Whether yaw/pitch/roll members are up-to-date since the last rotation matrix update.
double	m_yaw
double	m_pitch
double	m_roll
	These variables are updated every time that the object rotation matrix is modified (construction, loading from values, pose composition, etc )
RTTI stuff
typedef CPose3DPtr	SmartPtr
static mrpt::utils::CLASSINIT	_init_CPose3D
static mrpt::utils::TRuntimeClassId	classCPose3D
static const mrpt::utils::TRuntimeClassId *	classinfo
static const mrpt::utils::TRuntimeClassId *	_GetBaseClass ()
virtual const mrpt::utils::TRuntimeClassId *	GetRuntimeClass () const
	Returns information about the class of an object in runtime.
virtual mrpt::utils::CObject *	duplicate () const
	Returns a copy of the object, indepently of its class.
static mrpt::utils::CObject *	CreateObject ()
static CPose3DPtr	Create ()
Lie Algebra methods
void	ln (mrpt::math::CArrayDouble< 6 > &out_ln) const
	Take the logarithm of the 3x4 matrix defined by this pose, generating the corresponding vector in the SE(3) Lie Algebra.
mrpt::math::CArrayDouble< 6 >	ln () const
void	ln_jacob (mrpt::math::CMatrixFixedNumeric< double, 6, 12 > &J) const
	Jacobian of the logarithm of the 3x4 matrix defined by this pose.
CArrayDouble< 3 >	ln_rotation () const
	Take the logarithm of the 3x3 rotation matrix, generating the corresponding vector in the Lie Algebra.
static CPose3D	exp (const mrpt::math::CArrayNumeric< double, 6 > &vect)
	Exponentiate a Vector in the SE(3) Lie Algebra to generate a new CPose3D (static method).
static CMatrixDouble33	exp_rotation (const mrpt::math::CArrayNumeric< double, 3 > &vect)
	Exponentiate a vector in the Lie algebra to generate a new SO(3) (a 3x3 rotation matrix).
static void	ln_rot_jacob (const CMatrixDouble33 &R, CMatrixFixedNumeric< double, 3, 9 > &M)
	Static function to compute the Jacobian of the SO(3) Logarithm function, evaluated at a given 3x3 rotation matrix R.
STL-like methods and typedefs
enum	{ static_size = 6 }
typedef double	value_type
	The type of the elements.
typedef double &	reference
typedef const double &	const_reference
typedef std::size_t	size_type
typedef std::ptrdiff_t	difference_type
static size_type	size ()
static bool	empty ()
static size_type	max_size ()
static void	resize (const size_t n)
RTTI stuff
static const mrpt::utils::TRuntimeClassId	classCSerializable
class	mrpt::utils::CStream

Member Typedef Documentation

typedef const double& mrpt::poses::CPose3D::const_reference

Definition at line 479 of file CPose3D.h.

typedef std::ptrdiff_t mrpt::poses::CPose3D::difference_type

Definition at line 481 of file CPose3D.h.

typedef double& mrpt::poses::CPose3D::reference

Definition at line 478 of file CPose3D.h.

typedef std::size_t mrpt::poses::CPose3D::size_type

Definition at line 480 of file CPose3D.h.

typedef CPose3DPtr mrpt::poses::CPose3D::SmartPtr

A typedef for the associated smart pointer

Definition at line 75 of file CPose3D.h.

typedef CPose3D mrpt::poses::CPose3D::type_value

Used to emulate CPosePDF types, for example, in mrpt::graphs::CNetworkOfPoses.

Definition at line 465 of file CPose3D.h.

typedef double mrpt::poses::CPose3D::value_type

The type of the elements.

Definition at line 477 of file CPose3D.h.

Member Enumeration Documentation

anonymous enum

Enumerator:

is_3D_val

Definition at line 466 of file CPose3D.h.

anonymous enum

Enumerator:

rotation_dimensions

Definition at line 468 of file CPose3D.h.

anonymous enum

Enumerator:

is_PDF_val

Definition at line 469 of file CPose3D.h.

anonymous enum

Enumerator:

static_size

Definition at line 485 of file CPose3D.h.

Constructor & Destructor Documentation

mrpt::poses::CPose3D::CPose3D ( )

Default constructor, with all the coordinates set to zero.

mrpt::poses::CPose3D::CPose3D	(	const double	x,
		const double	y,
		const double	z,
		const double	yaw = `0`,
		const double	pitch = `0`,
		const double	roll = `0`
	)

Constructor with initilization of the pose; (remember that angles are always given in radians!)

mrpt::poses::CPose3D::CPose3D ( const math::CMatrixDouble & m ) [explicit]

Constructor from a 4x4 homogeneous matrix - the passed matrix can be actually of any size larger than or equal 3x4, since only those first values are used (the last row of a homogeneous 4x4 matrix are always fixed).

mrpt::poses::CPose3D::CPose3D ( const math::CMatrixDouble44 & m ) [explicit]

Constructor from a 4x4 homogeneous matrix:

template<class MATRIX33 , class VECTOR3 >

mrpt::poses::CPose3D::CPose3D	(	const MATRIX33 &	rot,
		const VECTOR3 &	xyz
	)		`[inline]`

Constructor from a 3x3 rotation matrix and a the translation given as a 3-vector, a 3-array, a CPoint3D or a TPoint3D.

Definition at line 109 of file CPose3D.h.

References ASSERT_EQUAL_, and mrpt::math::size().

mrpt::poses::CPose3D::CPose3D	(	const CMatrixDouble33 &	rot,
		const CArrayDouble< 3 > &	xyz
	)		`[inline]`

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 118 of file CPose3D.h.

mrpt::poses::CPose3D::CPose3D ( const CPose2D & )

Constructor from a CPose2D object.

mrpt::poses::CPose3D::CPose3D ( const CPoint3D & )

Constructor from a CPoint3D object.

mrpt::poses::CPose3D::CPose3D ( const mrpt::math::TPose3D & )

Constructor from lightweight object.

mrpt::poses::CPose3D::CPose3D	(	const mrpt::math::CQuaternionDouble &	q,
		const double	x,
		const double	y,
		const double	z
	)

Constructor from a quaternion (which only represents the 3D rotation part) and a 3D displacement.

mrpt::poses::CPose3D::CPose3D ( const CPose3DQuat & )

Constructor from a CPose3DQuat.

mrpt::poses::CPose3D::CPose3D ( TConstructorFlags_Poses constructor_dummy_param ) [inline]

Fast constructor that leaves all the data uninitialized - call with UNINITIALIZED_POSE as argument.

Definition at line 140 of file CPose3D.h.

mrpt::poses::CPose3D::CPose3D ( const CArrayDouble< 12 > & vec12 ) [inline, explicit]

Constructor from an array with these 12 elements: [r11 r21 r31 r12 r22 r32 r13 r23 r33 tx ty tz] where r{ij} are the entries of the 3x3 rotation matrix and t{x,y,z} are the 3D translation of the pose.

See also:: setFrom12Vector, getAs12Vector

Definition at line 146 of file CPose3D.h.

Member Function Documentation

static const mrpt::utils::TRuntimeClassId* mrpt::poses::CPose3D::_GetBaseClass ( ) [static, protected]

Reimplemented from mrpt::utils::CSerializable.

void mrpt::poses::CPose3D::addComponents ( const CPose3D & p )

Scalar sum of all 6 components: This is diferent from poses composition, which is implemented as "+" operators.

See also:: normalizeAngles

void mrpt::poses::CPose3D::asString ( std::string & s ) const [inline]

Returns a human-readable textual representation of the object (eg: "[x y z yaw pitch roll]", angles in degrees.)

See also:: fromString

Definition at line 403 of file CPose3D.h.

References mrpt::format(), and mrpt::utils::RAD2DEG().

std::string mrpt::poses::CPose3D::asString ( ) const [inline]

Definition at line 404 of file CPose3D.h.

References asString().

Referenced by asString().

void mrpt::poses::CPose3D::changeCoordinatesReference ( const CPose3D & p ) [inline]

makes: this = p (+) this

Definition at line 268 of file CPose3D.h.

CObject* mrpt::utils::CObject::clone ( ) const [inline, inherited]

Cloning interface for smart pointers.

Reimplemented in mrpt::opengl::CRenderizable, and mrpt::opengl::CRenderizableDisplayList.

Definition at line 154 of file CObject.h.

void mrpt::poses::CPose3D::composeFrom	(	const CPose3D &	A,
		const CPose3D &	B
	)

Makes "this = A (+) B"; this method is slightly more efficient than "this= A + B;" since it avoids the temporary object.

Note:: A or B can be "this" without problems.

Referenced by operator+(), mrpt::slam::PF_implementation::PF_SLAM_particlesEvaluator_AuxPFStandard(), and mrpt::slam::PF_implementation::PF_SLAM_aux_perform_one_rejection_sampling_step().

void mrpt::poses::CPose3D::composePoint	(	double	lx,
		double	ly,
		double	lz,
		double &	gx,
		double &	gy,
		double &	gz,
		mrpt::math::CMatrixFixedNumeric< double, 3, 3 > *	out_jacobian_df_dpoint = `NULL`,
		mrpt::math::CMatrixFixedNumeric< double, 3, 6 > *	out_jacobian_df_dpose = `NULL`,
		mrpt::math::CMatrixFixedNumeric< double, 3, 6 > *	out_jacobian_df_dse3 = `NULL`,
		bool	use_small_rot_approx = `false`
	)		const

An alternative, slightly more efficient way of doing $G = P \oplus L$ with G and L being 3D points and P this 6D pose.

If pointers are provided, the corresponding Jacobians are returned. "out_jacobian_df_dse3" stands for the Jacobian with respect to the 6D locally Euclidean vector in the tangent space of SE(3). See this report for mathematical details.

Parameters:

If	set to true, the Jacobian "out_jacobian_df_dpose" uses a fastest linearized appoximation (valid only for small rotations!).

Referenced by mrpt::math::project3D(), and mrpt::vision::pinhole::projectPoint_no_distortion().

void mrpt::poses::CPose3D::composePoint	(	const TPoint3D	local_point,
		TPoint3D &	global_point
	)		const `[inline]`

An alternative, slightly more efficient way of doing $G = P \oplus L$ with G and L being 3D points and P this 6D pose.

Note:: local_point is passed by value to allow global and local point to be the same variable

Definition at line 216 of file CPose3D.h.

References mrpt::math::TPoint3D::x, mrpt::math::TPoint3D::y, and mrpt::math::TPoint3D::z.

void mrpt::poses::CPose3D::composePoint	(	double	lx,
		double	ly,
		double	lz,
		float &	gx,
		float &	gy,
		float &	gz
	)		const `[inline]`

An alternative, slightly more efficient way of doing $G = P \oplus L$ with G and L being 3D points and P this 6D pose.

Definition at line 221 of file CPose3D.h.

static CPose3DPtr mrpt::poses::CPose3D::Create ( ) [static]

static mrpt::utils::CObject* mrpt::poses::CPose3D::CreateObject ( ) [static]

double mrpt::poses::CPose3D::distanceEuclidean6D ( const CPose3D & o ) const

The euclidean distance between two poses taken as two 6-length vectors (angles in radians).

virtual mrpt::utils::CObject* mrpt::poses::CPose3D::duplicate ( ) const [virtual]

Returns a copy of the object, indepently of its class.

Implements mrpt::utils::CObject.

mrpt::utils::CObjectPtr mrpt::utils::CObject::duplicateGetSmartPtr ( ) const [inline, inherited]

Returns a copy of the object, indepently of its class, as a smart pointer (the newly created object will exist as long as any copy of this smart pointer).

Definition at line 151 of file CObject.h.

static bool mrpt::poses::CPose3D::empty ( ) [inline, static]

Definition at line 487 of file CPose3D.h.

static CPose3D mrpt::poses::CPose3D::exp ( const mrpt::math::CArrayNumeric< double, 6 > & vect ) [static]

Exponentiate a Vector in the SE(3) Lie Algebra to generate a new CPose3D (static method).

Note:: Method from TooN (C) Tom Drummond (GNU GPL)

Referenced by mrpt::poses::SE_traits< 3 >::exp().

static CMatrixDouble33 mrpt::poses::CPose3D::exp_rotation ( const mrpt::math::CArrayNumeric< double, 3 > & vect ) [static]

Exponentiate a vector in the Lie algebra to generate a new SO(3) (a 3x3 rotation matrix).

Note:: Method from TooN (C) Tom Drummond (GNU GPL)

void mrpt::poses::CPose3D::fromString ( const std::string & s ) [inline]

Set the current object value from a string generated by 'asString' (eg: "[x y z yaw pitch roll]", angles in deg.

)

See also:: asString

Exceptions:

std::exception On invalid format

Definition at line 410 of file CPose3D.h.

References THROW_EXCEPTION, ASSERTMSG_, mrpt::math::size(), and mrpt::utils::DEG2RAD().

template<class ARRAYORVECTOR >

void mrpt::poses::CPose3D::getAs12Vector ( ARRAYORVECTOR & vec12 ) const [inline]

Get the pose representation as an array with these 12 elements: [r11 r21 r31 r12 r22 r32 r13 r23 r33 tx ty tz] where r{ij} are the entries of the 3x3 rotation matrix and t{x,y,z} are the 3D translation of the pose.

See also:: setFrom12Vector

Definition at line 349 of file CPose3D.h.

void mrpt::poses::CPose3D::getAsQuaternion	(	mrpt::math::CQuaternionDouble &	q,
		mrpt::math::CMatrixFixedNumeric< double, 4, 3 > *	out_dq_dr = `NULL`
	)		const

Returns the quaternion associated to the rotation of this object (NOTE: XYZ translation is ignored)

$\mathbf{q} = \left( \begin{array}{c} \cos (\phi /2) \cos (\theta /2) \cos (\psi /2) + \sin (\phi /2) \sin (\theta /2) \sin (\psi /2) \\ \sin (\phi /2) \cos (\theta /2) \cos (\psi /2) - \cos (\phi /2) \sin (\theta /2) \sin (\psi /2) \\ \cos (\phi /2) \sin (\theta /2) \cos (\psi /2) + \sin (\phi /2) \cos (\theta /2) \sin (\psi /2) \\ \cos (\phi /2) \cos (\theta /2) \sin (\psi /2) - \sin (\phi /2) \sin (\theta /2) \cos (\psi /2) \\ \end{array}\right)$

With : $\phi = roll$ , $\theta = pitch$ and $\psi = yaw$ .

Parameters:

out_dq_dr If provided, the 4x3 Jacobian of the transformation will be computed and stored here. It's the Jacobian of the transformation from (yaw pitch roll) to (qr qx qy qz).

Referenced by mrpt::math::jacobians::jacob_quat_from_yawpitchroll().

void mrpt::poses::CPose3D::getAsVector ( vector_double & v ) const

Returns a 1x6 vector with [x y z yaw pitch roll].

void mrpt::poses::CPose3D::getHomogeneousMatrix ( CMatrixDouble44 & out_HM ) const [inline]

Returns the corresponding 4x4 homogeneous transformation matrix for the point(translation) or pose (translation+orientation).

See also:: getInverseHomogeneousMatrix, getRotationMatrix

Definition at line 160 of file CPose3D.h.

CMatrixDouble44 mrpt::poses::CPose3D::getHomogeneousMatrixVal ( ) const [inline]

Definition at line 167 of file CPose3D.h.

const type_value& mrpt::poses::CPose3D::getPoseMean ( ) const [inline]

Definition at line 472 of file CPose3D.h.

type_value& mrpt::poses::CPose3D::getPoseMean ( ) [inline]

Definition at line 473 of file CPose3D.h.

void mrpt::poses::CPose3D::getRotationMatrix ( mrpt::math::CMatrixDouble33 & ROT ) const [inline]

Get the 3x3 rotation matrix.

See also:: getHomogeneousMatrix

Definition at line 170 of file CPose3D.h.

const mrpt::math::CMatrixDouble33& mrpt::poses::CPose3D::getRotationMatrix ( ) const [inline]

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 172 of file CPose3D.h.

virtual const mrpt::utils::TRuntimeClassId* mrpt::poses::CPose3D::GetRuntimeClass ( ) const [virtual]

Returns information about the class of an object in runtime.

Reimplemented from mrpt::utils::CSerializable.

void mrpt::poses::CPose3D::getYawPitchRoll	(	double &	yaw,
		double &	pitch,
		double &	roll
	)		const

Returns the three angles (yaw, pitch, roll), in radians, from the rotation matrix.

See also:: setFromValues, yaw, pitch, roll

void mrpt::poses::CPose3D::inverse ( )

Convert this pose into its inverse, saving the result in itself.

See also:: operator-

void mrpt::poses::CPose3D::inverseComposeFrom	(	const CPose3D &	A,
		const CPose3D &	B
	)

Makes $this = A \ominus B$ this method is slightly more efficient than "this= A - B;" since it avoids the temporary object.

Note:: A or B can be "this" without problems.

See also:: composeFrom, composePoint

Referenced by operator-().

void mrpt::poses::CPose3D::inverseComposePoint	(	const double	gx,
		const double	gy,
		const double	gz,
		double &	lx,
		double &	ly,
		double &	lz,
		mrpt::math::CMatrixFixedNumeric< double, 3, 3 > *	out_jacobian_df_dpoint = `NULL`,
		mrpt::math::CMatrixFixedNumeric< double, 3, 6 > *	out_jacobian_df_dpose = `NULL`,
		mrpt::math::CMatrixFixedNumeric< double, 3, 6 > *	out_jacobian_df_dse3 = `NULL`
	)		const

Computes the 3D point L such as $L = G \ominus this$ .

If pointers are provided, the corresponding Jacobians are returned. "out_jacobian_df_dse3" stands for the Jacobian with respect to the 6D locally Euclidean vector in the tangent space of SE(3). See this report for mathematical details.

See also:: composePoint, composeFrom

Referenced by mrpt::vision::pinhole::projectPoint_no_distortion().

static bool mrpt::poses::CPose3D::is_3D ( ) [inline, static]

Definition at line 467 of file CPose3D.h.

static bool mrpt::poses::CPose3D::is_PDF ( ) [inline, static]

Definition at line 470 of file CPose3D.h.

bool mrpt::poses::CPose3D::isHorizontal ( const double tolerance = 0 ) const

Return true if the 6D pose represents a Z axis almost exactly vertical (upwards or downwards), with a given tolerance (if set to 0 exact horizontality is tested).

void mrpt::poses::CPose3D::ln ( mrpt::math::CArrayDouble< 6 > & out_ln ) const

Take the logarithm of the 3x4 matrix defined by this pose, generating the corresponding vector in the SE(3) Lie Algebra.

Note:: Method from TooN (C) Tom Drummond (GNU GPL)

See also:: ln_jacob

Referenced by mrpt::poses::SE_traits< 3 >::ln().

mrpt::math::CArrayDouble<6> mrpt::poses::CPose3D::ln ( ) const [inline]

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 446 of file CPose3D.h.

References ln().

Referenced by ln().

void mrpt::poses::CPose3D::ln_jacob ( mrpt::math::CMatrixFixedNumeric< double, 6, 12 > & J ) const

Jacobian of the logarithm of the 3x4 matrix defined by this pose.

Note:: Method from TooN (C) Tom Drummond (GNU GPL)

See also:: ln

static void mrpt::poses::CPose3D::ln_rot_jacob	(	const CMatrixDouble33 &	R,
		CMatrixFixedNumeric< double, 3, 9 > &	M
	)		`[static]`

Static function to compute the Jacobian of the SO(3) Logarithm function, evaluated at a given 3x3 rotation matrix R.

See also:: ln, ln_jacob

CArrayDouble<3> mrpt::poses::CPose3D::ln_rotation ( ) const

Take the logarithm of the 3x3 rotation matrix, generating the corresponding vector in the Lie Algebra.

Note:: Method from TooN (C) Tom Drummond (GNU GPL)

static size_type mrpt::poses::CPose3D::max_size ( ) [inline, static]

Definition at line 488 of file CPose3D.h.

References static_size.

void mrpt::poses::CPose3D::normalizeAngles ( )

Rebuild the internal matrix & update the yaw/pitch/roll angles within the ]-PI,PI] range (Must be called after using addComponents)

See also:: addComponents

void mrpt::poses::CPose3D::operator*= ( const double s )

Scalar multiplication of x,y,z,yaw,pitch & roll (angles will be wrapped to the ]-pi,pi] interval).

CPose3D mrpt::poses::CPose3D::operator+ ( const CPose3D & b ) const [inline]

The operator $a \oplus b$ is the pose compounding operator.

Definition at line 181 of file CPose3D.h.

References mrpt::poses::UNINITIALIZED_POSE, and composeFrom().

CPoint3D mrpt::poses::CPose3D::operator+ ( const CPoint3D & b ) const

The operator $a \oplus b$ is the pose compounding operator.

CPoint3D mrpt::poses::CPose3D::operator+ ( const CPoint2D & b ) const

The operator $a \oplus b$ is the pose compounding operator.

CPose3D& mrpt::poses::CPose3D::operator+= ( const CPose3D & b ) [inline]

Make $this = this \oplus b$ (b can be "this" without problems)

Definition at line 244 of file CPose3D.h.

CPose3D mrpt::poses::CPose3D::operator- ( const CPose3D & b ) const [inline]

Compute $RET = this \oplus b$ .

Reimplemented from mrpt::poses::CPose< CPose3D >.

Definition at line 257 of file CPose3D.h.

References mrpt::poses::UNINITIALIZED_POSE, and inverseComposeFrom().

const double& mrpt::poses::CPose3D::operator[] ( unsigned int i ) const [inline]

Definition at line 381 of file CPose3D.h.

double mrpt::poses::CPose3D::pitch ( ) const [inline]

Get the PITCH angle (in radians)

See also:: setFromValues

Definition at line 365 of file CPose3D.h.

Referenced by mrpt::graphs::detail::graph_ops::write_VERTEX_line(), mrpt::graphs::detail::graph_ops::write_EDGE_line(), and mrpt::graphs::detail::graph_ops::auxEuclid2Dist().

virtual void mrpt::utils::CSerializable::readFromStream	(	mrpt::utils::CStream &	in,
		int	version
	)		`[protected, pure virtual, inherited]`

Introduces a pure virtual method responsible for loading from a CStream This can not be used directly be users, instead use "stream >> object;" for reading it from a stream or "stream >> object_ptr;" if the class is unknown apriori.

Parameters:

in	The input binary stream where the object data must read from.
version	The 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::exception On any error, see CStream::ReadBuffer

See also:: CStream

Implemented in mrpt::math::CMatrixD, and mrpt::math::CMatrix.

void mrpt::poses::CPose3D::rebuildRotationMatrix ( ) [protected]

Rebuild the homog matrix from the angles.

static void mrpt::poses::CPose3D::resize ( const size_t n ) [inline, static]

Definition at line 489 of file CPose3D.h.

References static_size, and mrpt::format().

double mrpt::poses::CPose3D::roll ( ) const [inline]

Get the ROLL angle (in radians)

See also:: setFromValues

Definition at line 366 of file CPose3D.h.

Referenced by mrpt::graphs::detail::graph_ops::write_VERTEX_line(), mrpt::graphs::detail::graph_ops::write_EDGE_line(), and mrpt::graphs::detail::graph_ops::auxEuclid2Dist().

template<class ARRAYORVECTOR >

void mrpt::poses::CPose3D::setFrom12Vector ( const ARRAYORVECTOR & vec12 ) [inline]

Set pose from an array with these 12 elements: [r11 r21 r31 r12 r22 r32 r13 r23 r33 tx ty tz] where r{ij} are the entries of the 3x3 rotation matrix and t{x,y,z} are the 3D translation of the pose.

See also:: getAs12Vector

Definition at line 334 of file CPose3D.h.

void mrpt::poses::CPose3D::setFromValues	(	const double	x0,
		const double	y0,
		const double	z0,
		const double	yaw = `0`,
		const double	pitch = `0`,
		const double	roll = `0`
	)

Set the pose from a 3D position (meters) and yaw/pitch/roll angles (radians) - This method recomputes the internal rotation matrix.

See also:: getYawPitchRoll, setYawPitchRoll

template<typename VECTORLIKE >

void mrpt::poses::CPose3D::setFromXYZQ	(	const VECTORLIKE &	v,
		const size_t	index_offset = `0`
	)		`[inline]`

Set the pose from a 3D position (meters) and a quaternion, stored as [x y z qr qx qy qz] in a 7-element vector.

See also:: setFromValues, getYawPitchRoll, setYawPitchRoll, CQuaternion, getAsQuaternion

Definition at line 304 of file CPose3D.h.

References ASSERT_ABOVEEQ_.

void mrpt::poses::CPose3D::setYawPitchRoll	(	const double	yaw_,
		const double	pitch_,
		const double	roll_
	)		`[inline]`

Set the 3 angles of the 3D pose (in radians) - This method recomputes the internal rotation coordinates matrix.

See also:: getYawPitchRoll, setFromValues

Definition at line 321 of file CPose3D.h.

References Eigen::internal::y.

static size_type mrpt::poses::CPose3D::size ( ) [inline, static]

Definition at line 486 of file CPose3D.h.

References static_size.

void mrpt::poses::CPose3D::sphericalCoordinates	(	const TPoint3D &	point,
		double &	out_range,
		double &	out_yaw,
		double &	out_pitch
	)		const

Computes the spherical coordinates of a 3D point as seen from the 6D pose specified by this object.

For the coordinate system see the top of this page.

void mrpt::poses::CPose3D::updateYawPitchRoll ( ) const [inline, protected]

Updates Yaw/pitch/roll members from the m_ROT.

Definition at line 89 of file CPose3D.h.

virtual void mrpt::utils::CSerializable::writeToStream	(	mrpt::utils::CStream &	out,
		int *	getVersion
	)		const `[protected, pure virtual, inherited]`

Introduces a pure virtual method responsible for writing to a CStream.

This can not be used directly be users, instead use "stream << object;" for writing it to a stream.

Parameters:

out	The output binary stream where object must be dumped.
getVersion	If NULL, the object must be dumped. If not, only the version of the object dump must be returned in this pointer. This enables the versioning of objects dumping and backward compatibility with previously stored data.

Exceptions:

std::exception On any error, see CStream::WriteBuffer