Bayesian Filtering Library Generated from SVN r
Public Member Functions | Protected Attributes
MeasurementModel< MeasVar, StateVar > Class Template Reference

#include <measurementmodel.h>

List of all members.

Public Member Functions

 MeasurementModel (ConditionalPdf< MeasVar, StateVar > *Measurementpdf=NULL)
 Constructor.
virtual ~MeasurementModel ()
 Destructor.
int MeasurementSizeGet () const
 Get Measurement Size.
bool SystemWithoutSensorParams () const
 Number of Conditional Arguments.
ConditionalPdf< MeasVar,
StateVar > * 
MeasurementPdfGet ()
 Get the MeasurementPDF.
void MeasurementPdfSet (ConditionalPdf< MeasVar, StateVar > *pdf)
 Set the MeasurementPDF.
MeasVar Simulate (const StateVar &x, const StateVar &s, int sampling_method=DEFAULT, void *sampling_args=NULL)
 Simulate the Measurement, given a certain state, and an input.
MeasVar Simulate (const StateVar &x, int sampling_method=DEFAULT, void *sampling_args=NULL)
 Simulate the system (no input system)
Probability ProbabilityGet (const MeasVar &z, const StateVar &x, const StateVar &s)
 Get the probability of a certain measurement.
Probability ProbabilityGet (const MeasVar &z, const StateVar &x)
 Get the probability of a certain measurement.

Protected Attributes

ConditionalPdf< MeasVar,
StateVar > * 
_MeasurementPdf
 ConditionalPdf representing $ P(Z_k | X_{k}, U_{k}) $.
bool _systemWithoutSensorParams
 System with no sensor params??

Detailed Description

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

Template class representing all possible (continu and discrete) Measurement Models

Todo:
Check if there should be a "model" base class...
Note:
Contrary to the system model, this template class has 2 template arguments: this is because of the different nature of the 2 conditional densities $ P ( Z | X ) $ and $ P ( X_k | X_{k-1} ) $ If $ X_{k-1} $ is discrete, then $ X_{k} $ will also be discrete, but a discrete state doesn't automatically imply a discrete measurement (as is proven in ASR!)

Definition at line 53 of file measurementmodel.h.


Constructor & Destructor Documentation

MeasurementModel ( ConditionalPdf< MeasVar, StateVar > *  Measurementpdf = NULL)

Constructor.

Parameters:
MeasurementpdfConditionalPdf<MeasVar,StateVar> representing $ P(Z_k | X_{k} (, U_{k})) $
See also:
MEASUREMENT_SIZE, STATE_SIZE, INPUT_SIZE, _MeasurementPdf

Definition at line 27 of file measurementmodel.h.


Member Function Documentation

void MeasurementPdfSet ( ConditionalPdf< MeasVar, StateVar > *  pdf)

Set the MeasurementPDF.

Parameters:
pdfa pointer to the measurement pdf

Definition at line 96 of file measurementmodel.h.

Probability ProbabilityGet ( const MeasVar &  z,
const StateVar &  x 
)

Get the probability of a certain measurement.

(measurement independent of input) gived a certain state and input

Parameters:
zthe measurement value
xx current state of the system
Returns:
the "probability" of the measurement

Definition at line 162 of file measurementmodel.h.

Probability ProbabilityGet ( const MeasVar &  z,
const StateVar &  x,
const StateVar &  s 
)

Get the probability of a certain measurement.

given a certain state and input

Parameters:
zthe measurement value
xcurrent state of the system
sthe sensor param value
Returns:
the "probability" of the measurement

Definition at line 151 of file measurementmodel.h.

MeasVar Simulate ( const StateVar &  x,
int  sampling_method = DEFAULT,
void *  sampling_args = NULL 
)

Simulate the system (no input system)

Parameters:
xcurrent state of the system
Returns:
State where we arrive by simulating the measurement model
Note:
Maybe the return value would better be a Sample<StateVar> instead of a StateVar
Parameters:
sampling_methodthe sampling method to be used while sampling from the Conditional Pdf describing the system (if not specified = DEFAULT)
sampling_argsSometimes a sampling method can have some extra parameters (eg mcmc sampling)

Definition at line 138 of file measurementmodel.h.

MeasVar Simulate ( const StateVar &  x,
const StateVar &  s,
int  sampling_method = DEFAULT,
void *  sampling_args = NULL 
)

Simulate the Measurement, given a certain state, and an input.

Parameters:
xcurrent state of the system
ssensor parameter
Returns:
Measurement generated by simulating the measurement model
Parameters:
sampling_methodthe sampling method to be used while sampling from the Conditional Pdf describing the system (if not specified = DEFAULT)
sampling_argsSometimes a sampling method can have some extra parameters (eg mcmc sampling)
Note:
Maybe the return value would better be a Sample<StateVar> instead of a StateVar

Definition at line 122 of file measurementmodel.h.


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