BFL::Filter< StateVar, MeasVar > Class Template Reference

Abstract class representing an interface for Bayesian Filters. More...

#include <filter.h>

Inheritance diagram for BFL::Filter< StateVar, MeasVar >:

List of all members.

Public Member Functions
	Filter (Pdf< StateVar > *prior)
	Constructor.
	Filter (const Filter< StateVar, MeasVar > &filt)
	copy constructor
virtual Pdf< StateVar > *	PostGet ()
	Get Posterior density.
virtual void	Reset (Pdf< StateVar > *prior)
	Reset Filter.
int	TimeStepGet () const
	Get current time.
virtual bool	Update (SystemModel< StateVar > *const sysmodel, const StateVar &u, MeasurementModel< MeasVar, StateVar > *const measmodel, const MeasVar &z, const StateVar &s)
	Full Update (system with inputs/sensing params)
virtual bool	Update (SystemModel< StateVar > *const sysmodel, MeasurementModel< MeasVar, StateVar > *const measmodel, const MeasVar &z, const StateVar &s)
	Full Update (system without inputs, with sensing params)
virtual bool	Update (SystemModel< StateVar > *const sysmodel, MeasurementModel< MeasVar, StateVar > *const measmodel, const MeasVar &z)
	Full Update (system without inputs/sensing params)
virtual bool	Update (SystemModel< StateVar > *const sysmodel, const StateVar &u, MeasurementModel< MeasVar, StateVar > *const measmodel, const MeasVar &z)
	Full Update (system with inputs, without sensing params)
virtual bool	Update (SystemModel< StateVar > *const sysmodel, const StateVar &u)
	System Update (system with inputs)
virtual bool	Update (SystemModel< StateVar > *const sysmodel)
	System Update (system without inputs)
virtual bool	Update (MeasurementModel< MeasVar, StateVar > *const measmodel, const MeasVar &z, const StateVar &s)
	Measurement Update (system with "sensing params")
virtual bool	Update (MeasurementModel< MeasVar, StateVar > *const measmodel, const MeasVar &z)
	Measurement Update (system without "sensing params")
virtual	~Filter ()
	destructor
Protected Member Functions
virtual bool	UpdateInternal (SystemModel< StateVar > *const sysmodel, const StateVar &u, MeasurementModel< MeasVar, StateVar > *const measmodel, const MeasVar &z, const StateVar &s)=0
	Actual implementation of Update, varies along filters.
Protected Attributes
Pdf< StateVar > *	_post
	Pointer to the Posterior Pdf.
Pdf< StateVar > *	_prior
	prior Pdf
int	_timestep
	Represents the current timestep of the filter.

---

Detailed Description

template<typename StateVar, typename MeasVar>
class BFL::Filter< StateVar, MeasVar >

Abstract class representing an interface for Bayesian Filters.

This is the Abstract interface class that defines the interface of Bayesian filters. These filters are all related to i) a System Model, ii) a Measurement Model and iii) a Prior density reflecting the subjective belief of the person applying the filter BEFORE getting sensor or any other form of information about the modeled system.

This class is the base class for particle filters, kalman filters, ...

This class is a template class with 2 templates. In this way it allows filtering for "semi-discrete" models, eg. models with a fixed number of states (discrete states) but with continuous observations, as needed in Automatic Speech Recognition.

See also:

Pdf SystemModel MeasurementModel ConditionalPdf

Bug:

For now, due to a "bug" (= non-existence of a feature :-) in the ConditionalPdf class, STATES AND INPUTS MUST BE OF THE SAME TYPE (both discrete, or both continuous! This means that you can use this class for the following model types:

• States, inputs and measurements continuous (most frequently used?)
• States and inputs continous, Measurements discrete
• States and inputs discrete, Measurements continous
• States, inputs and measurements discrete

StateVar represents the nature of the states and inputs MeasVar represents the nature of the measurements

BEWARE: The order of the template arguments is reversed with respect to the notation used in "measurementmodel.h"

Definition at line 77 of file filter.h.

---

Constructor & Destructor Documentation

template<typename SVar, typename MVar >

BFL::Filter< SVar, MVar >::Filter

(

Pdf< SVar > *

prior

)

Constructor.

Precondition:

you created the prior

Parameters:

prior

pointer to the prior Pdf

Definition at line 27 of file filter.h.

template<typename SVar, typename MVar>

BFL::Filter< SVar, MVar >::Filter

(

const Filter< SVar, MVar > &

filt

)

copy constructor

Bug:

we should make a copy of the prior

Todo:

Check if we should make a copy of the pdf's too?

Bug:

we should make a copy of the pdf's too

Definition at line 38 of file filter.h.

template<typename SVar , typename MVar >

BFL::Filter< SVar, MVar >::~Filter

(

)

[virtual]

destructor

Definition at line 32 of file filter.cpp.

---

Member Function Documentation

template<typename SVar , typename MVar >

Pdf< SVar > * BFL::Filter< SVar, MVar >::PostGet

(

)

[virtual]

Get Posterior density.

Get the current Posterior density

Returns:

a pointer to the current posterior

Reimplemented in BFL::MixtureParticleFilter< StateVar, MeasVar >, BFL::ParticleFilter< StateVar, MeasVar >, BFL::ParticleFilter< ColumnVector, ColumnVector >, BFL::HistogramFilter< MeasVar >, and BFL::KalmanFilter.

Definition at line 126 of file filter.cpp.

template<typename SVar, typename MVar >

void BFL::Filter< SVar, MVar >::Reset

(

Pdf< SVar > *

prior

)

[virtual]

Reset Filter.

Definition at line 41 of file filter.cpp.

template<typename SVar , typename MVar >

int BFL::Filter< SVar, MVar >::TimeStepGet

(

)

const

Get current time.

Get the current time of the filter

Returns:

the current timestep

Definition at line 49 of file filter.cpp.

template<typename SVar, typename MVar>

bool BFL::Filter< SVar, MVar >::Update	(	SystemModel< SVar > *const	sysmodel,
		const SVar &	u,
		MeasurementModel< MVar, SVar > *const	measmodel,
		const MVar &	z,
		const SVar &	s
	)		[virtual]

Full Update (system with inputs/sensing params)

Parameters:

sysmodel	pointer to the system model to use for update
u	input to the system
measmodel	pointer to the measurement model to use for update
z	measurement
s	"sensing parameter"

Definition at line 56 of file filter.h.

template<typename SVar, typename MVar>

bool BFL::Filter< SVar, MVar >::Update	(	SystemModel< SVar > *const	sysmodel,
		MeasurementModel< MVar, SVar > *const	measmodel,
		const MVar &	z,
		const SVar &	s
	)		[virtual]

Full Update (system without inputs, with sensing params)

Parameters:

sysmodel	pointer to the system model to use for update
measmodel	pointer to the measurement model to use for update
z	measurement
s	"sensing parameter"

Definition at line 66 of file filter.h.

template<typename SVar, typename MVar>

bool BFL::Filter< SVar, MVar >::Update	(	SystemModel< SVar > *const	sysmodel,
		MeasurementModel< MVar, SVar > *const	measmodel,
		const MVar &	z
	)		[virtual]

Full Update (system without inputs/sensing params)

Parameters:

sysmodel	pointer to the system model to use for update
measmodel	pointer to the measurement model to use for update
z	measurement

Definition at line 76 of file filter.h.

template<typename SVar, typename MVar>

bool BFL::Filter< SVar, MVar >::Update	(	SystemModel< SVar > *const	sysmodel,
		const SVar &	u,
		MeasurementModel< MVar, SVar > *const	measmodel,
		const MVar &	z
	)		[virtual]

Full Update (system with inputs, without sensing params)

Parameters:

sysmodel	pointer to the system model to use for update
u	input to the system
measmodel	pointer to the measurement model to use for update
z	measurement

Definition at line 85 of file filter.h.

template<typename SVar, typename MVar >

bool BFL::Filter< SVar, MVar >::Update	(	SystemModel< SVar > *const	sysmodel,
		const SVar &	u
	)		[virtual]

System Update (system with inputs)

Parameters:

sysmodel	pointer to the system model to use for update
u	input to the system

Definition at line 95 of file filter.h.

template<typename SVar, typename MVar >

bool BFL::Filter< SVar, MVar >::Update

(

SystemModel< SVar > *const

sysmodel

)

[virtual]

System Update (system without inputs)

Parameters:

sysmodel

pointer to the system model to use for update

Definition at line 103 of file filter.h.

template<typename SVar, typename MVar>

bool BFL::Filter< SVar, MVar >::Update	(	MeasurementModel< MVar, SVar > *const	measmodel,
		const MVar &	z,
		const SVar &	s
	)		[virtual]

Measurement Update (system with "sensing params")

Parameters:

measmodel	pointer to the measurement model to use for update
z	measurement
s	"sensing parameter"

Definition at line 110 of file filter.h.

template<typename SVar, typename MVar>

bool BFL::Filter< SVar, MVar >::Update	(	MeasurementModel< MVar, SVar > *const	measmodel,
		const MVar &	z
	)		[virtual]

Measurement Update (system without "sensing params")

Parameters:

measmodel	pointer to the measurement model to use for update
z	measurement

Definition at line 119 of file filter.h.

template<typename StateVar, typename MeasVar>

virtual bool BFL::Filter< StateVar, MeasVar >::UpdateInternal	(	SystemModel< StateVar > *const	sysmodel,
		const StateVar &	u,
		MeasurementModel< MeasVar, StateVar > *const	measmodel,
		const MeasVar &	z,
		const StateVar &	s
	)		[protected, pure virtual]

Actual implementation of Update, varies along filters.

Parameters:

sysmodel	pointer to the used system model
u	input param for proposal density
measmodel	pointer to the used measurementmodel
z	measurement param for proposal density
s	sensor param for proposal density

Implemented in BFL::ParticleFilter< StateVar, MeasVar >, BFL::KalmanFilter, BFL::MixtureParticleFilter< StateVar, MeasVar >, BFL::HistogramFilter< MeasVar >, BFL::ASIRFilter< StateVar, MeasVar >, BFL::BootstrapFilter< StateVar, MeasVar >, and BFL::MixtureBootstrapFilter< StateVar, MeasVar >.

---

Member Data Documentation

template<typename StateVar, typename MeasVar>

Pdf<StateVar>* BFL::Filter< StateVar, MeasVar >::_post [protected]

Pointer to the Posterior Pdf.

The Posterior Pdf represents the subjective belief of the person applying the filter AFTER processing inputs and measurements. A filter does not maintain the beliefs at all timesteps t, since this leads to non-constant (or ever growing if you prefer) memory requirements. However, it is possible, to copy the Posterior density at all timesteps in your application by means of the PostGet() member function

See also:

PostGet()

Definition at line 95 of file filter.h.

template<typename StateVar, typename MeasVar>

Pdf<StateVar>* BFL::Filter< StateVar, MeasVar >::_prior [protected]

prior Pdf

Definition at line 82 of file filter.h.

template<typename StateVar, typename MeasVar>

int BFL::Filter< StateVar, MeasVar >::_timestep [protected]

Represents the current timestep of the filter.

Todo:

Check wether this really belongs here

Definition at line 100 of file filter.h.

---

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

• filter.h
• filter.cpp