gov.sandia.cognition.statistics.bayesian

Interface ParticleFilter<ObservationType,ParameterType>

Type Parameters:

ObservationType - Type of observations handled by the algorithm.

ParameterType - Type of parameters to infer.

All Superinterfaces:

BatchLearner<java.util.Collection<? extends ObservationType>,DataDistribution<ParameterType>>, BayesianEstimator<ObservationType,ParameterType,DataDistribution<ParameterType>>, java.lang.Cloneable, CloneableSerializable, IncrementalLearner<ObservationType,DataDistribution<ParameterType>>, Randomized, RecursiveBayesianEstimator<ObservationType,ParameterType,DataDistribution<ParameterType>>, java.io.Serializable

All Known Implementing Classes:

AbstractParticleFilter, SamplingImportanceResamplingParticleFilter

@PublicationReference(author={"M. Sanjeev Arulampalam","Simon Maskell","Neil Gordon","Tim Clapp"},title="A Tutorial on Particle Filters for Online Nonlinear/Non-Gaussian Bayesian Tracking",type=Journal,publication="IEEE Transactions on Signal Processing, Vol. 50, No. 2",year=2002,pages={174,188},url="http://people.cs.ubc.ca/~murphyk/Software/Kalman/ParticleFilterTutorial.pdf") @PublicationReference(author="Wikipedia",title="Particle filter",type=WebPage,year=2009,url="http://en.wikipedia.org/wiki/Particle_filter")
public interface ParticleFilter<ObservationType,ParameterType>
extends RecursiveBayesianEstimator<ObservationType,ParameterType,DataDistribution<ParameterType>>, Randomized

A particle filter aims to estimate a sequence of hidden parameters based on observed data using point-mass estimates of the posterior distribution. Particle filters are sometimes called Sequential Monte Carlo estimation.

Since:

3.0

Author:

Kevin R. Dixon

Nested Class Summary

Nested Classes

Modifier and Type	Interface and Description
static interface	ParticleFilter.Updater<ObservationType,ParameterType> Updates the particles.

Method Summary

Modifier and Type	Method and Description
double	computeEffectiveParticles(DataDistribution<ParameterType> particles) Computes the effective number of particles.
int	getNumParticles() Gets the number of particles
ParticleFilter.Updater<ObservationType,ParameterType>	getUpdater() Gets the updater
void	setNumParticles(int numParticles) Sets the number of particles

Methods inherited from interface gov.sandia.cognition.learning.algorithm.BatchLearner

learn

Methods inherited from interface gov.sandia.cognition.util.CloneableSerializable

clone

Methods inherited from interface gov.sandia.cognition.learning.algorithm.IncrementalLearner

createInitialLearnedObject, update, update

Methods inherited from interface gov.sandia.cognition.util.Randomized

getRandom, setRandom

Method Detail

getUpdater

ParticleFilter.Updater<ObservationType,ParameterType> getUpdater()

Gets the updater

Returns:

Updater algorithm that updates the particles.

getNumParticles

int getNumParticles()

Gets the number of particles

Returns:

Number of particles.

setNumParticles

void setNumParticles(int numParticles)

Sets the number of particles

Parameters:

numParticles - Number of particles.

computeEffectiveParticles

double computeEffectiveParticles(DataDistribution<ParameterType> particles)

Computes the effective number of particles.

Parameters:

particles - Current state of the Particle filter.

Returns:

Effective number of particles.