Bayesian Filtering Library Generated from SVN r
Classes | Public Member Functions | Protected Member Functions | Protected Attributes | Friends
IteratedExtendedKalmanFilter Class Reference

#include <iteratedextendedkalmanfilter.h>

Inheritance diagram for IteratedExtendedKalmanFilter:
KalmanFilter Filter< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector >

List of all members.

Classes

struct  MeasUpdateVariablesIExt

Public Member Functions

 IteratedExtendedKalmanFilter (Gaussian *prior, unsigned int nr_it=1, InnovationCheck *innov=NULL)
virtual ~IteratedExtendedKalmanFilter ()
 Destructor.
void AllocateMeasModelIExt (const vector< unsigned int > &meas_dimensions)
 Function to allocate memory needed during the measurement update,.
void AllocateMeasModelIExt (const unsigned int &meas_dimensions)
 Function to allocate memory needed during the measurement update.
virtual GaussianPostGet ()
 Get Posterior density.
void AllocateMeasModel (const vector< unsigned int > &meas_dimensions)
 Function to allocate memory needed during the measurement update,.
void AllocateMeasModel (const unsigned int &meas_dimensions)
 Function to allocate memory needed during the measurement update.
virtual void Reset (Pdf< MatrixWrapper::ColumnVector > *prior)
 Reset Filter.
virtual bool Update (SystemModel< MatrixWrapper::ColumnVector > *const sysmodel, const MatrixWrapper::ColumnVector &u, MeasurementModel< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector > *const measmodel, const MatrixWrapper::ColumnVector &z, const MatrixWrapper::ColumnVector &s)
 Full Update (system with inputs/sensing params)
virtual bool Update (SystemModel< MatrixWrapper::ColumnVector > *const sysmodel, MeasurementModel< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector > *const measmodel, const MatrixWrapper::ColumnVector &z, const MatrixWrapper::ColumnVector &s)
 Full Update (system without inputs, with sensing params)
virtual bool Update (SystemModel< MatrixWrapper::ColumnVector > *const sysmodel, MeasurementModel< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector > *const measmodel, const MatrixWrapper::ColumnVector &z)
 Full Update (system without inputs/sensing params)
virtual bool Update (SystemModel< MatrixWrapper::ColumnVector > *const sysmodel, const MatrixWrapper::ColumnVector &u, MeasurementModel< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector > *const measmodel, const MatrixWrapper::ColumnVector &z)
 Full Update (system with inputs, without sensing params)
virtual bool Update (SystemModel< MatrixWrapper::ColumnVector > *const sysmodel, const MatrixWrapper::ColumnVector &u)
 System Update (system with inputs)
virtual bool Update (SystemModel< MatrixWrapper::ColumnVector > *const sysmodel)
 System Update (system without inputs)
virtual bool Update (MeasurementModel< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector > *const measmodel, const MatrixWrapper::ColumnVector &z, const MatrixWrapper::ColumnVector &s)
 Measurement Update (system with "sensing params")
virtual bool Update (MeasurementModel< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector > *const measmodel, const MatrixWrapper::ColumnVector &z)
 Measurement Update (system without "sensing params")
int TimeStepGet () const
 Get current time.

Protected Member Functions

virtual void SysUpdate (SystemModel< MatrixWrapper::ColumnVector > *const sysmodel, const MatrixWrapper::ColumnVector &u)
 System Update.
virtual void MeasUpdate (MeasurementModel< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector > *const measmodel, const MatrixWrapper::ColumnVector &z, const MatrixWrapper::ColumnVector &s)
 Measurement Update (overloaded)
void PostSigmaSet (const MatrixWrapper::SymmetricMatrix &s)
 Set covariance of posterior estimate.
void PostMuSet (const MatrixWrapper::ColumnVector &c)
 Set expected value of posterior estimate.
void CalculateSysUpdate (const MatrixWrapper::ColumnVector &J, const MatrixWrapper::Matrix &F, const MatrixWrapper::SymmetricMatrix &Q)
void CalculateMeasUpdate (const MatrixWrapper::ColumnVector &z, const MatrixWrapper::ColumnVector &Z, const MatrixWrapper::Matrix &H, const MatrixWrapper::SymmetricMatrix &R)
virtual bool UpdateInternal (SystemModel< MatrixWrapper::ColumnVector > *const sysmodel, const MatrixWrapper::ColumnVector &u, MeasurementModel< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector > *const measmodel, const MatrixWrapper::ColumnVector &z, const MatrixWrapper::ColumnVector &s)
 Actual implementation of Update, varies along filters.

Protected Attributes

ColumnVector _Mu_new
SymmetricMatrix _Sigma_new
Matrix _Sigma_temp
Matrix _Sigma_temp_par
Matrix _SMatrix
Matrix _K
std::map< unsigned int,
MeasUpdateVariables > 
_mapMeasUpdateVariables
std::map< unsigned int,
MeasUpdateVariables >
::iterator 
_mapMeasUpdateVariables_it
Pdf
< MatrixWrapper::ColumnVector > * 
_prior
 prior Pdf
Pdf
< MatrixWrapper::ColumnVector > * 
_post
 Pointer to the Posterior Pdf.
int _timestep
 Represents the current timestep of the filter.

Friends

class NonminimalKalmanFilter

Detailed Description

This is a class implementing the Kalman Filter (KF) class for Iterated Extended Kalman Filters.

The System- and MeasurementUpdate equasions are not linear, and will be approximated by local linearisations.

See also:
KalmanFilter

Definition at line 40 of file iteratedextendedkalmanfilter.h.


Constructor & Destructor Documentation

IteratedExtendedKalmanFilter ( Gaussian prior,
unsigned int  nr_it = 1,
InnovationCheck innov = NULL 
)

Constructor

Precondition:
you created the prior
Parameters:
priorpointer to the Monte Carlo Pdf prior density
nr_itthe number of iterations in one update
innovpointer to InnovationCheck (to end the iterations of the innovation is too small)

Member Function Documentation

void CalculateMeasUpdate ( const MatrixWrapper::ColumnVector &  z,
const MatrixWrapper::ColumnVector &  Z,
const MatrixWrapper::Matrix &  H,
const MatrixWrapper::SymmetricMatrix &  R 
) [protected, inherited]

Calculate Kalman filter Measurement Update

\[ x_k = x_{k-} + K.(z - Z) \]

\[ P_k = (I-K.H).P_{k-} \]

with

\[ K = P_{k-}.H'.(H.P_{k-}.H'+R)^{-1} \]

void CalculateSysUpdate ( const MatrixWrapper::ColumnVector &  J,
const MatrixWrapper::Matrix &  F,
const MatrixWrapper::SymmetricMatrix &  Q 
) [protected, inherited]

Calculate Kalman filter System Update

\[ x_k = J \]

\[ P_k = F.P_{k-}.F' + Q \]

virtual void MeasUpdate ( MeasurementModel< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector > *const  measmodel,
const MatrixWrapper::ColumnVector &  z,
const MatrixWrapper::ColumnVector &  s 
) [protected, virtual]

Measurement Update (overloaded)

Update the filter's Posterior density using the sensor measurements, an input and the measurement model. This method is used when the measurements depend on the inputs too (doesn't happen very often, does it?) BEWARE: the first time the measurment update is called with a new size of measurement, new allocations are done

Parameters:
measmodelpointer to the measurement model the filter should use
zsensor measurement
sinput to the system (must be of the same type as u for now, since this was not yet implemented in ConditionalPdf

Implements KalmanFilter.

virtual Gaussian* PostGet ( ) [virtual, inherited]

Get Posterior density.

Get the current Posterior density

Returns:
a pointer to the current posterior

Reimplemented from Filter< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector >.

virtual void SysUpdate ( SystemModel< MatrixWrapper::ColumnVector > *const  sysmodel,
const MatrixWrapper::ColumnVector &  u 
) [protected, virtual]

System Update.

Update the filter's Posterior density using the deterministic inputs to the system and the system model

Parameters:
sysmodelpointer to the system model the filter should use
uinput to the system

Implements KalmanFilter.

int TimeStepGet ( ) const [inherited]

Get current time.

Get the current time of the filter

Returns:
the current timestep
virtual bool Update ( SystemModel< MatrixWrapper::ColumnVector > *const  sysmodel) [virtual, inherited]

System Update (system without inputs)

Parameters:
sysmodelpointer to the system model to use for update
virtual bool Update ( SystemModel< MatrixWrapper::ColumnVector > *const  sysmodel,
MeasurementModel< MatrixWrapper::ColumnVector , MatrixWrapper::ColumnVector > *const  measmodel,
const MatrixWrapper::ColumnVector &  z 
) [virtual, inherited]

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
virtual bool Update ( SystemModel< MatrixWrapper::ColumnVector > *const  sysmodel,
const MatrixWrapper::ColumnVector &  u,
MeasurementModel< MatrixWrapper::ColumnVector , MatrixWrapper::ColumnVector > *const  measmodel,
const MatrixWrapper::ColumnVector &  z,
const MatrixWrapper::ColumnVector &  s 
) [virtual, inherited]

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"
virtual bool Update ( MeasurementModel< MatrixWrapper::ColumnVector , MatrixWrapper::ColumnVector > *const  measmodel,
const MatrixWrapper::ColumnVector &  z,
const MatrixWrapper::ColumnVector &  s 
) [virtual, inherited]

Measurement Update (system with "sensing params")

Parameters:
measmodelpointer to the measurement model to use for update
zmeasurement
s"sensing parameter"
virtual bool Update ( SystemModel< MatrixWrapper::ColumnVector > *const  sysmodel,
const MatrixWrapper::ColumnVector &  u,
MeasurementModel< MatrixWrapper::ColumnVector , MatrixWrapper::ColumnVector > *const  measmodel,
const MatrixWrapper::ColumnVector &  z 
) [virtual, inherited]

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
virtual bool Update ( SystemModel< MatrixWrapper::ColumnVector > *const  sysmodel,
const MatrixWrapper::ColumnVector &  u 
) [virtual, inherited]

System Update (system with inputs)

Parameters:
sysmodelpointer to the system model to use for update
uinput to the system
virtual bool Update ( SystemModel< MatrixWrapper::ColumnVector > *const  sysmodel,
MeasurementModel< MatrixWrapper::ColumnVector , MatrixWrapper::ColumnVector > *const  measmodel,
const MatrixWrapper::ColumnVector &  z,
const MatrixWrapper::ColumnVector &  s 
) [virtual, inherited]

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"
virtual bool Update ( MeasurementModel< MatrixWrapper::ColumnVector , MatrixWrapper::ColumnVector > *const  measmodel,
const MatrixWrapper::ColumnVector &  z 
) [virtual, inherited]

Measurement Update (system without "sensing params")

Parameters:
measmodelpointer to the measurement model to use for update
zmeasurement
virtual bool UpdateInternal ( SystemModel< MatrixWrapper::ColumnVector > *const  sysmodel,
const MatrixWrapper::ColumnVector &  u,
MeasurementModel< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector > *const  measmodel,
const MatrixWrapper::ColumnVector &  z,
const MatrixWrapper::ColumnVector &  s 
) [protected, virtual, inherited]

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< MatrixWrapper::ColumnVector, MatrixWrapper::ColumnVector >.


Friends And Related Function Documentation

friend class NonminimalKalmanFilter [friend, inherited]

Very dirty hack to avoid ugly methods PostSigmaSet and PostMuSet to be public! NonMinimalKalmanFilter should be redesigned though!

Definition at line 110 of file kalmanfilter.h.


Member Data Documentation

Pdf<MatrixWrapper::ColumnVector >* _post [protected, inherited]

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.

int _timestep [protected, inherited]

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: