ASIR: Auxiliary Particle Filter. More...

#include <asirfilter.h>

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

Public Member Functions
	ASIRFilter (MCPdf< StateVar > *prior, int resampleperiod=0, double resamplethreshold=0, int resamplescheme=DEFAULT_RS)
	Constructor.
virtual	~ASIRFilter ()
	Destructor.
Protected Member Functions
virtual void	UpdateInternal (SystemModel< StateVar > const sysmodel, const StateVar &u, MeasurementModel< MeasVar, StateVar > const measmodel, const MeasVar &z, const StateVar &s)
	Actual implementation of updateinternal.

Detailed Description

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

ASIR: Auxiliary Particle Filter.

This is a possible implementation of a particle filter, which in some/many cases will yield better results, since measures are taken to make the proposal density more similar to the posterior. See

      @Article{	  pitt_auxiliary,
      author	= {Pitt, M. and Shephard, N.},
      title	= {Filtering via simulation: auxiliary particle filter},
      journal	= {Journal of the American Statistical Association},
      year	= {1999},
      note	= {forthcoming}
      }

for more details.

Note that this particular implementation:

Uses currently a particular version of important sampling
Still uses the the system pdf as "actual" proposal density, which amongs others means the results will be suboptimal in case of a log concave measurement density.

Particular issue with of proposalstep in case of ASIR filter... The current implementation uses the approximation (see p. 11 of the Pitt and Shephard paper, we use their notation here ---

$\alpha$

denoting the state and

$y$

denoting the measurement)

$f(y_{t+1} | \alpha_{t+1}) \approx f(y_{t+1} | \mu_{t+1})$

where

$\mu$

is the mean value of the system pdf.

Note that the ASIR needs the measurementmodel for its proposalstep, to obtain a better proposal. Note also that normally, the ASIR performs better that the standard SIR filter in case of outliers, but worse for "normal data" (due to the extra resampling stage).

Definition at line 79 of file asirfilter.h.

Constructor & Destructor Documentation

template<typename StateVar , typename MeasVar >

BFL::ASIRFilter< SVar, MVar >::ASIRFilter	(	MCPdf< StateVar > *	prior,
		int	resampleperiod = `0`,
		double	resamplethreshold = `0`,
		int	resamplescheme = `DEFAULT_RS`
	)

Constructor.

Precondition:: you created the necessary models and the prior

Parameters:

prior	pointer to the Monte Carlo Pdf prior density
resampleperiod	fixed resampling period (if desired)
resamplethreshold	threshold used when dynamic resampling
resamplescheme	resampling scheme, see header file for different defines and their meaning

Definition at line 26 of file asirfilter.cpp.

template<typename SVar , typename MVar >

BFL::ASIRFilter< SVar, MVar >::~ASIRFilter ( ) [virtual]

Destructor.

Definition at line 44 of file asirfilter.cpp.

Member Function Documentation

template<typename StateVar , typename MeasVar >

void BFL::ASIRFilter< SVar, MVar >::UpdateInternal	(	SystemModel< StateVar > *const	sysmodel,
		const StateVar &	u,
		MeasurementModel< MeasVar, StateVar > *const	measmodel,
		const MeasVar &	z,
		const StateVar &	s
	)		`[protected, virtual]`

Actual implementation of updateinternal.

Reimplemented from BFL::ParticleFilter< StateVar, MeasVar >.

Definition at line 48 of file asirfilter.cpp.

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

Public Member Functions

Protected Member Functions

Detailed Description

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

Constructor & Destructor Documentation

Member Function Documentation

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