Class for linear analytic systemmodels with additive gaussian noise. More...

#include <linearanalyticsystemmodel_gaussianuncertainty.h>

Inheritance diagram for LinearAnalyticSystemModelGaussianUncertainty:

Public Member Functions
	LinearAnalyticSystemModelGaussianUncertainty (LinearAnalyticConditionalGaussian *pdf)
	Constructor. More...

virtual	~LinearAnalyticSystemModelGaussianUncertainty ()
	Destructor.

void	ASet (const MatrixWrapper::Matrix &a)
	Set Matrix A. More...

void	BSet (const MatrixWrapper::Matrix &b)
	Set Matrix B. More...

const MatrixWrapper::Matrix &	AGet () const
	Get Matrix A.

const MatrixWrapper::Matrix &	BGet () const
	Get Matrix B.

virtual MatrixWrapper::Matrix	df_dxGet (const MatrixWrapper::ColumnVector &u, const MatrixWrapper::ColumnVector &x)
	Returns F-matrix. More...

virtual MatrixWrapper::ColumnVector	PredictionGet (const MatrixWrapper::ColumnVector &u, const MatrixWrapper::ColumnVector &x)
	Returns prediction of state.

virtual MatrixWrapper::SymmetricMatrix	CovarianceGet (const MatrixWrapper::ColumnVector &u, const MatrixWrapper::ColumnVector &x)
	Covariance of system noise.

int	StateSizeGet () const
	Get State Size. More...

bool	SystemWithoutInputs () const
	Has the system inputs or not.

ConditionalPdf< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector > *	SystemPdfGet ()
	Get the SystemPDF. More...

void	SystemPdfSet (ConditionalPdf< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector > *pdf)
	Set the SystemPDF. More...

MatrixWrapper::ColumnVector	Simulate (const MatrixWrapper::ColumnVector &x, const MatrixWrapper::ColumnVector &u, const SampleMthd sampling_method=SampleMthd::DEFAULT, void *sampling_args=NULL)
	Simulate the system. More...

MatrixWrapper::ColumnVector	Simulate (const MatrixWrapper::ColumnVector &x, const SampleMthd sampling_method=SampleMthd::DEFAULT, void *sampling_args=NULL)
	Simulate the system (no input system) More...

Probability	ProbabilityGet (const MatrixWrapper::ColumnVector &x_k, const MatrixWrapper::ColumnVector &x_kminusone, const MatrixWrapper::ColumnVector &u)
	Get the probability of arriving in a next state. More...

Probability	ProbabilityGet (const MatrixWrapper::ColumnVector &x_k, const MatrixWrapper::ColumnVector &x_kminusone)
	Get the probability of arriving in a next state. More...

Protected Attributes
ConditionalPdf< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector > *	_SystemPdf
	ConditionalPdf representing $P(X_k \| X_{k-1}, U_{k})$ . More...

bool	_systemWithoutInputs
	System with no inputs? More...

Detailed Description

Class for linear analytic systemmodels with additive gaussian noise.

This class represents all systemmodels of the form

$x_k = A \times x_{k-1} + B \times u_{k} + N(\mu,\Sigma)$

Definition at line 33 of file linearanalyticsystemmodel_gaussianuncertainty.h.

Constructor & Destructor Documentation

◆ LinearAnalyticSystemModelGaussianUncertainty()

LinearAnalyticSystemModelGaussianUncertainty ( LinearAnalyticConditionalGaussian * pdf )

Constructor.

Precondition: LinearAnalyticConditionalGaussian should have 1/2 conditional Arguments (checked) and the first conditional argument should be x!

Parameters

pdf	Conditional pdf with Gaussian uncertainty

Member Function Documentation

◆ ASet()

void ASet ( const MatrixWrapper::Matrix & a )

Set Matrix A.

This can be particularly useful for time-varying systems

Parameters

a Matrix a

◆ BSet()

void BSet ( const MatrixWrapper::Matrix & b )

Set Matrix B.

This can be particularly useful for time-varying systems

Parameters

b Matrix b

◆ df_dxGet()

virtual MatrixWrapper::Matrix df_dxGet	(	const MatrixWrapper::ColumnVector &	u,
		const MatrixWrapper::ColumnVector &	x
	)

virtualinherited

Returns F-matrix.

$F = \frac{df}{dx} \mid_{u,x}$

used by kalman filter variants

Parameters

u	The value of the input in which the derivate is evaluated
x	The value in the state in which the derivate is evaluated

Bug:: Should actually be defined for any continuous system model! There should be a class between this one and system model tout court, not assuming gaussian uncertainty!

◆ ProbabilityGet() [1/2]

Probability ProbabilityGet	(	const MatrixWrapper::ColumnVector &	x_k,
		const MatrixWrapper::ColumnVector &	x_kminusone
	)

inherited

Get the probability of arriving in a next state.

(no-input-system)

Parameters

x_k	the next state (at time k)
x_kminusone	the current state (at time k-1)

Returns: the probability value

◆ ProbabilityGet() [2/2]

Probability ProbabilityGet	(	const MatrixWrapper::ColumnVector &	x_k,
		const MatrixWrapper::ColumnVector &	x_kminusone,
		const MatrixWrapper::ColumnVector &	u
	)

inherited

Get the probability of arriving in a next state.

Parameters

x_k	the next state (at time k)
x_kminusone	the current state (at time k-1)
u	the input

Returns: the probability value

◆ Simulate() [1/2]

MatrixWrapper::ColumnVector Simulate	(	const MatrixWrapper::ColumnVector &	x,
		const MatrixWrapper::ColumnVector &	u,
		const SampleMthd	sampling_method = `SampleMthd::DEFAULT`,
		void *	sampling_args = `NULL`
	)

inherited

Simulate the system.

Parameters

x	current state of the system
u	input to the system

Returns: State where we arrive by simulating the system model for 1 step

Parameters

sampling_method	the sampling method to be used while sampling from the Conditional Pdf describing the system (if not specified = DEFAULT)
sampling_args	Sometimes a sampling method can have some extra parameters (eg mcmc sampling)

Note: Maybe the return value would better be a Sample<T> instead of a T

◆ Simulate() [2/2]

MatrixWrapper::ColumnVector Simulate	(	const MatrixWrapper::ColumnVector &	x,
		const SampleMthd	sampling_method = `SampleMthd::DEFAULT`,
		void *	sampling_args = `NULL`
	)

inherited

Simulate the system (no input system)

Parameters

x	current state of the system

Returns: State where we arrive by simulating the system model for 1 step

Note: Maybe the return value would better be a Sample<T> instead of a T

Parameters

sampling_method	the sampling method to be used while sampling from the Conditional Pdf describing the system (if not specified = DEFAULT)
sampling_args	Sometimes a sampling method can have some extra parameters (eg mcmc sampling)

◆ StateSizeGet()

int StateSizeGet ( ) const

inherited

Get State Size.

Copy constructor SystemModel(const SystemModel<T>& model);

Returns: the statesize of the system

◆ SystemPdfGet()

ConditionalPdf< MatrixWrapper::ColumnVector , MatrixWrapper::ColumnVector > * SystemPdfGet ( )

inherited

Get the SystemPDF.

Returns: a reference to the ConditionalPdf describing the system

◆ SystemPdfSet()

void SystemPdfSet ( ConditionalPdf< MatrixWrapper::ColumnVector , MatrixWrapper::ColumnVector > * pdf )

inherited

Set the SystemPDF.

Parameters

pdf	a reference to the ConditionalPdf describing the system

Member Data Documentation

◆ _SystemPdf

ConditionalPdf<MatrixWrapper::ColumnVector ,MatrixWrapper::ColumnVector >* _SystemPdf

protectedinherited

ConditionalPdf representing $P(X_k | X_{k-1}, U_{k})$ .

Definition at line 59 of file systemmodel.h.

◆ _systemWithoutInputs

bool _systemWithoutInputs

protectedinherited

System with no inputs?

Definition at line 62 of file systemmodel.h.

The documentation for this class was generated from the following file:

linearanalyticsystemmodel_gaussianuncertainty.h

Public Member Functions

Protected Attributes

Detailed Description

Constructor & Destructor Documentation

◆ LinearAnalyticSystemModelGaussianUncertainty()

Member Function Documentation

◆ ASet()

◆ BSet()

◆ df_dxGet()

◆ ProbabilityGet() [1/2]

◆ ProbabilityGet() [2/2]

◆ Simulate() [1/2]

◆ Simulate() [2/2]

◆ StateSizeGet()

◆ SystemPdfGet()

◆ SystemPdfSet()

Member Data Documentation

◆ _SystemPdf

◆ _systemWithoutInputs