opentl::models::Motion Class Reference

Abstract class for object motion models. More...

Inherited by opentl::models::LinearARModels.

List of all members.

Public Types
Public Member Functions
virtual MotionType	getMotionType ()
	get motion type (see above)
virtual math::SquareMatrix &	getNoiseCovariance ()=0
	Get process noise covariance matrix.
virtual int	getNoiseDim ()=0
	Returns dimension of process noise square matrix.
virtual void	getUniformPrior (opentl::core::State &srcDstState)=0
	Function to get the first state prior for the tracking. The filter calls this function at initialization. Requires to call setNoiseCovariance() first to get range values for uniform distribution.
	Motion (MotionType motionType)
virtual void	predictMotion (opentl::core::State &srcDstState, bool addNoise=false, double deltaT=-1.0, math::SquareMatrix *noiseCovMat=NULL, math::SquareMatrix *jacStateTrans=NULL, math::Vector *inputU=NULL)=0
	Predicts state(k+1) based on specific motion model.
virtual void	setAveragePose (const math::Vector &averagePose)=0
	Set average pose value (used by some models with additive update).
virtual void	setDeltaT (double deltaT)
	Sets a fixed delta t (used if no one is provided at prediction time).
virtual void	setNoiseCovariance (const math::SquareMatrix &noiseCovMat)=0
	Set process noise covariance (needs to update matrices afterwards!).
virtual void	setUniformNoiseRange (const math::Vector &uniformNoiseRange)=0
	Set uniform noise range (for the initial prior).
virtual	~Motion ()
Protected Attributes
double	mDeltaT
	Use a fixed time value for deltaT (default: 1.0 sec).
bool	mDirtyFlag
	Dirty flag: if we change mDeltaT, then we need to update matrices.
MotionType	mMotionType
	motion type (see above)

---

Detailed Description

Abstract class for object motion models.

Author:

Claus Lenz, lenz@in.tum.de; Erwin Roth, erwin.roth@weihenstephan.org; Giorgio Panin, panin@in.tum.de

---

Member Enumeration Documentation

enum opentl::models::Motion::MotionType

ENUM type specifying the motion type (useful for quickly knowing the number of state derivatives or replica).

Enumerator:

BROWNIAN
CWNA
OSCILL

---

Constructor & Destructor Documentation

opentl::models::Motion::Motion

(

MotionType

motionType

)

virtual opentl::models::Motion::~Motion

(

)

[virtual]

---

Member Function Documentation

virtual MotionType opentl::models::Motion::getMotionType

(

)

[inline, virtual]

get motion type (see above)

virtual math::SquareMatrix& opentl::models::Motion::getNoiseCovariance

(

)

[pure virtual]

Get process noise covariance matrix.

Implemented in opentl::models::LinearARModels.

virtual int opentl::models::Motion::getNoiseDim

(

)

[pure virtual]

Returns dimension of process noise square matrix.

Returns:

Dimension of process noise

Implemented in opentl::models::LinearARModels.

virtual void opentl::models::Motion::getUniformPrior

(

opentl::core::State &

srcDstState

)

[pure virtual]

Function to get the first state prior for the tracking. The filter calls this function at initialization. Requires to call setNoiseCovariance() first to get range values for uniform distribution.

Parameters:

srcDstState

State used as basis for initialization of motion

Implemented in opentl::models::LinearARModels.

virtual void opentl::models::Motion::predictMotion	(	opentl::core::State &	srcDstState,
		bool	addNoise = false,
		double	deltaT = -1.0,
		math::SquareMatrix *	noiseCovMat = NULL,
		math::SquareMatrix *	jacStateTrans = NULL,
		math::Vector *	inputU = NULL
	)			[pure virtual]

Predicts state(k+1) based on specific motion model.

Parameters:

	srcDstState	Variable holding input and output state
	addNoise	Optional: (default = false) if true, process noise will be added to state(k+1)
	deltaT	Real time interval for the prediction [sec] (-1 = use fixed deltaT, internally set)
	noiseCovMat	Gives access to (dynamic) process noise covariance matrix
	jacStateTrans	Optional: returns Jacobian of state transition matrix (must be square)
	inputU	Optional: input control vector (u)

Implemented in opentl::models::LinearARModels.

virtual void opentl::models::Motion::setAveragePose

(

const math::Vector &

averagePose

)

[pure virtual]

Set average pose value (used by some models with additive update).

Parameters:

averagePose

Input average pose vector

Implemented in opentl::models::LinearARModels.

virtual void opentl::models::Motion::setDeltaT

(

double

deltaT

)

[inline, virtual]

Sets a fixed delta t (used if no one is provided at prediction time).

Parameters:

deltaT

value of delta t

virtual void opentl::models::Motion::setNoiseCovariance

(

const math::SquareMatrix &

noiseCovMat

)

[pure virtual]

Set process noise covariance (needs to update matrices afterwards!).

Implemented in opentl::models::LinearARModels.

virtual void opentl::models::Motion::setUniformNoiseRange

(

const math::Vector &

uniformNoiseRange

)

[pure virtual]

Set uniform noise range (for the initial prior).

Parameters:

uniformNoiseRange

Input minimum and maximum values for the absolute pose parameters

Implemented in opentl::models::LinearARModels.

---

Member Data Documentation

double opentl::models::Motion::mDeltaT [protected]

Use a fixed time value for deltaT (default: 1.0 sec).

bool opentl::models::Motion::mDirtyFlag [protected]

Dirty flag: if we change mDeltaT, then we need to update matrices.

MotionType opentl::models::Motion::mMotionType [protected]

motion type (see above)