Bayesian Filtering Library Generated from SVN r
Public Member Functions | Protected Member Functions | Protected Attributes | List of all members
MixtureParticleFilter< StateVar, MeasVar > Class Template Reference

Virtual Class representing all Mixture particle filters. More...

#include <mixtureParticleFilter.h>

Inheritance diagram for MixtureParticleFilter< StateVar, MeasVar >:
Filter< StateVar, MeasVar > MixtureBootstrapFilter< StateVar, MeasVar >

Public Member Functions

 MixtureParticleFilter (Mixture< StateVar > *prior, ConditionalPdf< StateVar, StateVar > *proposal, int resampleperiod=0, double resamplethreshold=0, int resamplescheme=DEFAULT_RS, int maintainMixturePeriod=1)
 Constructor.
 
 MixtureParticleFilter (Mixture< StateVar > *prior, Mixture< StateVar > *post, ConditionalPdf< StateVar, StateVar > *proposal, int resampleperiod=0, double resamplethreshold=0, int resamplescheme=DEFAULT_RS, int maintainMixturePeriod=1)
 Constructor.
 
virtual ~MixtureParticleFilter ()
 Destructor.
 
 MixtureParticleFilter (const MixtureParticleFilter< StateVar, MeasVar > &filt)
 Copy Constructor.
 
virtual void Reset (Mixture< StateVar > *prior)
 Reset Filter.
 
virtual void ProposalSet (ConditionalPdf< StateVar, StateVar > *const cpdf)
 Set the proposal density.
 
ConditionalPdf< StateVar, StateVar > * ProposalGet ()
 Get a pointer to the proposal density.
 
virtual Mixture< StateVar > * PostGet ()
 Get Posterior density.
 
virtual void Reset (Pdf< StateVar > *prior)
 Reset Filter.
 
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")
 
int TimeStepGet () const
 Get current time.
 

Protected Member Functions

virtual bool UpdateInternal (SystemModel< StateVar > *const sysmodel, const StateVar &u, MeasurementModel< MeasVar, StateVar > *const measmodel, const MeasVar &z, const StateVar &s)
 Actual implementation of Update, varies along filters.
 
virtual bool ProposalStepInternal (SystemModel< StateVar > *const sysmodel, const StateVar &u, MeasurementModel< MeasVar, StateVar > *const measmodel, const MeasVar &z, const StateVar &s)
 Proposal step.
 
virtual bool ProposalStepInternalOne (int component, SystemModel< StateVar > *const sysmodel, const StateVar &u, MeasurementModel< MeasVar, StateVar > *const measmodel, const MeasVar &z, const StateVar &s)
 Proposal step for one component.
 
virtual bool UpdateWeightsInternal (SystemModel< StateVar > *const sysmodel, const StateVar &u, MeasurementModel< MeasVar, StateVar > *const measmodel, const MeasVar &z, const StateVar &s)
 Update Weights.
 
virtual bool UpdateWeightsInternalOne (int component, SystemModel< StateVar > *const sysmodel, const StateVar &u, MeasurementModel< MeasVar, StateVar > *const measmodel, const MeasVar &z, const StateVar &s)
 Update Weights for one component.
 
virtual bool DynamicResampleStep ()
 Resample if necessary.
 
virtual bool DynamicResampleStepOne (int component)
 Resampling for one component.
 
virtual bool StaticResampleStep ()
 Resample if wanted.
 
virtual bool Resample ()
 Actual Resampling happens here;.
 
virtual bool ResampleOne (int component)
 Actual Resampling for one component;.
 
virtual bool MaintainMixtureStep ()
 Maintain Mixture if wanted.
 
virtual bool MaintainMixture ()
 Actual mixture maintainance happens here;.
 

Protected Attributes

ConditionalPdf< StateVar, StateVar > * _proposal
 Pointer to the Proposal Density.
 
WeightedSample< StateVar > _sample
 While updating use sample<StateVar>
 
vector< vector< WeightedSample< StateVar > > > _old_samplesVec
 While updating store list of old samples.
 
vector< vector< WeightedSample< StateVar > > > _new_samplesVec
 While updating store list of new samples.
 
vector< vector< Sample< StateVar > > > _new_samples_unweightedVec
 While resampling.
 
vector< WeightedSample< StateVar > >::iterator _os_it
 Iterator for old list of samples.
 
vector< WeightedSample< StateVar > >::iterator _ns_it
 Iterator for new list of samples.
 
vector< Probability_newMixtureWeights
 Vector containing the new mixture weights during update step.
 
vector< Probability_sumWeights
 Vector containing the sum of weights during update step.
 
int _resamplePeriod
 Number of timestep between resampling from the Posterior Pdf.
 
double _resampleThreshold
 Threshold used when dynamic resampling.
 
int _resampleScheme
 Which resample algorithm (see top of particle.h for #defines)
 
bool _dynamicResampling
 Dynamic resampling or fixed period resampling?
 
bool _proposal_depends_on_meas
 Proposal depends on last measurement?
 
bool _created_post
 created own post
 
int _maintainMixturePeriod
 Number of timestep between mixture maintainance of the Posterior.
 
Pdf< StateVar > * _prior
 prior Pdf
 
Pdf< StateVar > * _post
 Pointer to the Posterior Pdf.
 
int _timestep
 Represents the current timestep of the filter.
 

Detailed Description

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

Virtual Class representing all Mixture particle filters.

This is a virtual class representing the family of all mixture particle filters. Mixture Particle filters are filters in which the Posterior density is represented by a mixture (class Mixture) of Monte Carlo Pdfs (class MCPdf).

However, the updating of the Posterior density can still be done in several ways, that's why the System and Measurement update members are still pure virtual functions.

This class is the base class for all sorts of mixture particle filters.

See also
Mixture
MCPdf
Sample
WeightedSample
Bug
Resampling is not implemented generically enough yet. There's only the possibility to choose between static period resampling and dynamic resampling as proposed by Jun Liu. The correct way of implementing this would be to create a virtual function that has to be implemented by the user, but this creates more hassle for the user (a different particle filter for each scheme).

Definition at line 153 of file mixtureParticleFilter.h.

Constructor & Destructor Documentation

◆ MixtureParticleFilter() [1/3]

template<typename StateVar , typename MeasVar >
MixtureParticleFilter ( Mixture< StateVar > * prior,
ConditionalPdf< StateVar, StateVar > * proposal,
int resampleperiod = 0,
double resamplethreshold = 0,
int resamplescheme = DEFAULT_RS,
int maintainMixturePeriod = 1 )

Constructor.

Precondition
you created the necessary models and the prior
Parameters
priorpointer to the mixture of Monte Carlo Pdfs prior density
proposalpointer to the proposal density to use
resampleperiodfixed resampling period (if desired)
resamplethresholdthreshold used when dynamic resampling
resampleschemeresampling scheme, see header file for different defines and their meaning
maintainMixturePeriodfixed mixture maintainance period
Bug

prior should be of type pdf and not mcpdf. See also notes with implementation

let the user implement her/his own resamplescheme

◆ MixtureParticleFilter() [2/3]

template<typename StateVar , typename MeasVar >
MixtureParticleFilter ( Mixture< StateVar > * prior,
Mixture< StateVar > * post,
ConditionalPdf< StateVar, StateVar > * proposal,
int resampleperiod = 0,
double resamplethreshold = 0,
int resamplescheme = DEFAULT_RS,
int maintainMixturePeriod = 1 )

Constructor.

Precondition
you created the necessary models and the prior
Parameters
priorpointer to the mixture of Monte Carlo Pdfs prior density
postpointer to the mixture of Monte Carlo Pdfs post density
proposalpointer to the proposal density to use
resampleperiodfixed resampling period (if desired)
resamplethresholdthreshold used when dynamic resampling
resampleschemeresampling scheme, see header file for different defines and their meaning
maintainMixturePeriodfixed mixture maintainance period
Bug

prior should be of type pdf and not mcpdf. See also notes with implementation

let the user implement her/his own resamplescheme

◆ MixtureParticleFilter() [3/3]

template<typename StateVar , typename MeasVar >
MixtureParticleFilter ( const MixtureParticleFilter< StateVar, MeasVar > & filt)

Copy Constructor.

Bug
implementation probably contains a bug

Member Function Documentation

◆ DynamicResampleStepOne()

template<typename StateVar , typename MeasVar >
virtual bool DynamicResampleStepOne ( int component)
protectedvirtual

Resampling for one component.

Parameters
componentcomponent number of component to update
Bug
let the user implement her/his own resamplescheme

◆ MaintainMixtureStep()

template<typename StateVar , typename MeasVar >
virtual bool MaintainMixtureStep ( )
protectedvirtual

Maintain Mixture if wanted.

Bug
let the user implement her/his own mixture maintaince scheme

◆ PostGet()

template<typename StateVar , typename MeasVar >
virtual Mixture< StateVar > * PostGet ( )
virtual

Get Posterior density.

Get the current Posterior density

Returns
a pointer to the current posterior

Reimplemented from Filter< StateVar, MeasVar >.

◆ ProposalGet()

template<typename StateVar , typename MeasVar >
ConditionalPdf< StateVar, StateVar > * ProposalGet ( )

Get a pointer to the proposal density.

Returns
a pointer to the proposal density

◆ ProposalSet()

template<typename StateVar , typename MeasVar >
virtual void ProposalSet ( ConditionalPdf< StateVar, StateVar > *const cpdf)
virtual

Set the proposal density.

Parameters
cpdfthe new proposal density. The order of the conditional arguments is fixed and should be: x (state), u (input), z (measurement), s (sensor param). Off course all of them are optional

◆ ProposalStepInternal()

template<typename StateVar , typename MeasVar >
virtual bool ProposalStepInternal ( SystemModel< StateVar > *const sysmodel,
const StateVar & u,
MeasurementModel< MeasVar, StateVar > *const measmodel,
const MeasVar & z,
const StateVar & s )
protectedvirtual

Proposal step.

Implementation of proposal step

Parameters
sysmodelpointer to the used system model
uinput param for proposal density
measmodelpointer to the used measurementmodel
zmeasurement param for proposal density
ssensor param for proposal density
Bug
Make sampling method variable. See implementation.

◆ ProposalStepInternalOne()

template<typename StateVar , typename MeasVar >
virtual bool ProposalStepInternalOne ( int component,
SystemModel< StateVar > *const sysmodel,
const StateVar & u,
MeasurementModel< MeasVar, StateVar > *const measmodel,
const MeasVar & z,
const StateVar & s )
protectedvirtual

Proposal step for one component.

Implementation of proposal step

Parameters
componentcomponent number of component to update
sysmodelpointer to the used system model
uinput param for proposal density
measmodelpointer to the used measurementmodel
zmeasurement param for proposal density
ssensor param for proposal density
Bug
Make sampling method variable. See implementation.

◆ ResampleOne()

template<typename StateVar , typename MeasVar >
virtual bool ResampleOne ( int component)
protectedvirtual

Actual Resampling for one component;.

Parameters
componentcomponent number of component to update

◆ StaticResampleStep()

template<typename StateVar , typename MeasVar >
virtual bool StaticResampleStep ( )
protectedvirtual

Resample if wanted.

Bug
let the user implement her/his own resamplescheme

◆ TimeStepGet()

template<typename StateVar , typename MeasVar >
int TimeStepGet ( ) const
inherited

Get current time.

Get the current time of the filter

Returns
the current timestep

◆ Update() [1/8]

template<typename StateVar , typename MeasVar >
virtual bool Update ( MeasurementModel< MeasVar, StateVar > *const measmodel,
const MeasVar & z )
virtualinherited

Measurement Update (system without "sensing params")

Parameters
measmodelpointer to the measurement model to use for update
zmeasurement

Reimplemented in Optimalimportancefilter< StateVar, MeasVar >.

◆ Update() [2/8]

template<typename StateVar , typename MeasVar >
virtual bool Update ( MeasurementModel< MeasVar, StateVar > *const measmodel,
const MeasVar & z,
const StateVar & s )
virtualinherited

Measurement Update (system with "sensing params")

Parameters
measmodelpointer to the measurement model to use for update
zmeasurement
s"sensing parameter"

Reimplemented in Optimalimportancefilter< StateVar, MeasVar >.

◆ Update() [3/8]

template<typename StateVar , typename MeasVar >
virtual bool Update ( SystemModel< StateVar > *const sysmodel)
virtualinherited

System Update (system without inputs)

Parameters
sysmodelpointer to the system model to use for update

Reimplemented in Optimalimportancefilter< StateVar, MeasVar >.

◆ Update() [4/8]

template<typename StateVar , typename MeasVar >
virtual bool Update ( SystemModel< StateVar > *const sysmodel,
const StateVar & u )
virtualinherited

System Update (system with inputs)

Parameters
sysmodelpointer to the system model to use for update
uinput to the system

Reimplemented in Optimalimportancefilter< StateVar, MeasVar >.

◆ Update() [5/8]

template<typename StateVar , typename MeasVar >
virtual bool Update ( SystemModel< StateVar > *const sysmodel,
const StateVar & u,
MeasurementModel< MeasVar, StateVar > *const measmodel,
const MeasVar & z )
virtualinherited

Full Update (system with inputs, without sensing params)

Parameters
sysmodelpointer to the system model to use for update
uinput to the system
measmodelpointer to the measurement model to use for update
zmeasurement

Reimplemented in Optimalimportancefilter< StateVar, MeasVar >.

◆ Update() [6/8]

template<typename StateVar , typename MeasVar >
virtual bool Update ( SystemModel< StateVar > *const sysmodel,
const StateVar & u,
MeasurementModel< MeasVar, StateVar > *const measmodel,
const MeasVar & z,
const StateVar & s )
virtualinherited

Full Update (system with inputs/sensing params)

Parameters
sysmodelpointer to the system model to use for update
uinput to the system
measmodelpointer to the measurement model to use for update
zmeasurement
s"sensing parameter"

Reimplemented in Optimalimportancefilter< StateVar, MeasVar >.

◆ Update() [7/8]

template<typename StateVar , typename MeasVar >
virtual bool Update ( SystemModel< StateVar > *const sysmodel,
MeasurementModel< MeasVar, StateVar > *const measmodel,
const MeasVar & z )
virtualinherited

Full Update (system without inputs/sensing params)

Parameters
sysmodelpointer to the system model to use for update
measmodelpointer to the measurement model to use for update
zmeasurement

Reimplemented in Optimalimportancefilter< StateVar, MeasVar >.

◆ Update() [8/8]

template<typename StateVar , typename MeasVar >
virtual bool Update ( SystemModel< StateVar > *const sysmodel,
MeasurementModel< MeasVar, StateVar > *const measmodel,
const MeasVar & z,
const StateVar & s )
virtualinherited

Full Update (system without inputs, with sensing params)

Parameters
sysmodelpointer to the system model to use for update
measmodelpointer to the measurement model to use for update
zmeasurement
s"sensing parameter"

Reimplemented in Optimalimportancefilter< StateVar, MeasVar >.

◆ UpdateInternal()

template<typename StateVar , typename MeasVar >
virtual bool UpdateInternal ( SystemModel< StateVar > *const sysmodel,
const StateVar & u,
MeasurementModel< MeasVar, StateVar > *const measmodel,
const MeasVar & z,
const StateVar & s )
protectedvirtual

Actual implementation of Update, varies along filters.

Parameters
sysmodelpointer to the used system model
uinput param for proposal density
measmodelpointer to the used measurementmodel
zmeasurement param for proposal density
ssensor param for proposal density

Implements Filter< StateVar, MeasVar >.

Reimplemented in MixtureBootstrapFilter< StateVar, MeasVar >.

◆ UpdateWeightsInternal()

template<typename StateVar , typename MeasVar >
virtual bool UpdateWeightsInternal ( SystemModel< StateVar > *const sysmodel,
const StateVar & u,
MeasurementModel< MeasVar, StateVar > *const measmodel,
const MeasVar & z,
const StateVar & s )
protectedvirtual

Update Weights.

Parameters
sysmodelpointer to the used system model
uinput param for proposal density
measmodelpointer to the used measurementmodel
zmeasurement param for proposal density
ssensor param for proposal density

◆ UpdateWeightsInternalOne()

template<typename StateVar , typename MeasVar >
virtual bool UpdateWeightsInternalOne ( int component,
SystemModel< StateVar > *const sysmodel,
const StateVar & u,
MeasurementModel< MeasVar, StateVar > *const measmodel,
const MeasVar & z,
const StateVar & s )
protectedvirtual

Update Weights for one component.

Parameters
componentcomponent number of component to update
sysmodelpointer to the used system model
uinput param for proposal density
measmodelpointer to the used measurementmodel
zmeasurement param for proposal density
ssensor param for proposal density

Member Data Documentation

◆ _created_post

template<typename StateVar , typename MeasVar >
bool _created_post
protected

created own post

Definition at line 206 of file mixtureParticleFilter.h.

◆ _dynamicResampling

template<typename StateVar , typename MeasVar >
bool _dynamicResampling
protected

Dynamic resampling or fixed period resampling?

Definition at line 200 of file mixtureParticleFilter.h.

◆ _maintainMixturePeriod

template<typename StateVar , typename MeasVar >
int _maintainMixturePeriod
protected

Number of timestep between mixture maintainance of the Posterior.

By choosing this period, one can execute period maintainance of mixtures. You can implement maintainMixture such that target splitting, appearing and disappearing is handled

Definition at line 213 of file mixtureParticleFilter.h.

◆ _new_samples_unweightedVec

template<typename StateVar , typename MeasVar >
vector< vector<Sample<StateVar> > > _new_samples_unweightedVec
protected

While resampling.

Definition at line 177 of file mixtureParticleFilter.h.

◆ _new_samplesVec

template<typename StateVar , typename MeasVar >
vector<vector<WeightedSample<StateVar> > > _new_samplesVec
protected

While updating store list of new samples.

Definition at line 175 of file mixtureParticleFilter.h.

◆ _newMixtureWeights

template<typename StateVar , typename MeasVar >
vector<Probability> _newMixtureWeights
protected

Vector containing the new mixture weights during update step.

Definition at line 183 of file mixtureParticleFilter.h.

◆ _ns_it

template<typename StateVar , typename MeasVar >
vector<WeightedSample<StateVar>>::iterator _ns_it
protected

Iterator for new list of samples.

Definition at line 181 of file mixtureParticleFilter.h.

◆ _old_samplesVec

template<typename StateVar , typename MeasVar >
vector<vector<WeightedSample<StateVar> > > _old_samplesVec
protected

While updating store list of old samples.

Definition at line 173 of file mixtureParticleFilter.h.

◆ _os_it

template<typename StateVar , typename MeasVar >
vector<WeightedSample<StateVar>>::iterator _os_it
protected

Iterator for old list of samples.

Definition at line 179 of file mixtureParticleFilter.h.

◆ _post

template<typename StateVar , typename MeasVar >
Pdf<StateVar>* _post
protectedinherited

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.

◆ _prior

template<typename StateVar , typename MeasVar >
Pdf<StateVar>* _prior
protectedinherited

prior Pdf

Definition at line 82 of file filter.h.

◆ _proposal

template<typename StateVar , typename MeasVar >
ConditionalPdf<StateVar,StateVar>* _proposal
protected

Pointer to the Proposal Density.

Every particle filter (or more correct: every Sequential Importance Sampling method) uses a proposal density to do the forward sampling step

Definition at line 168 of file mixtureParticleFilter.h.

◆ _proposal_depends_on_meas

template<typename StateVar , typename MeasVar >
bool _proposal_depends_on_meas
protected

Proposal depends on last measurement?

Definition at line 203 of file mixtureParticleFilter.h.

◆ _resamplePeriod

template<typename StateVar , typename MeasVar >
int _resamplePeriod
protected

Number of timestep between resampling from the Posterior Pdf.

By choosing this period, one can avoid numerical instability (aka Degeneration of the particle filter

Definition at line 191 of file mixtureParticleFilter.h.

◆ _resampleScheme

template<typename StateVar , typename MeasVar >
int _resampleScheme
protected

Which resample algorithm (see top of particle.h for #defines)

Definition at line 197 of file mixtureParticleFilter.h.

◆ _resampleThreshold

template<typename StateVar , typename MeasVar >
double _resampleThreshold
protected

Threshold used when dynamic resampling.

Definition at line 194 of file mixtureParticleFilter.h.

◆ _sample

template<typename StateVar , typename MeasVar >
WeightedSample<StateVar> _sample
protected

While updating use sample<StateVar>

Definition at line 171 of file mixtureParticleFilter.h.

◆ _sumWeights

template<typename StateVar , typename MeasVar >
vector<Probability> _sumWeights
protected

Vector containing the sum of weights during update step.

Definition at line 185 of file mixtureParticleFilter.h.

◆ _timestep

template<typename StateVar , typename MeasVar >
int _timestep
protectedinherited

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 file: