A class template for Kalman filtering. More...

#include <cf_filters.h>

Inheritance diagram for pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >:

Collaboration diagram for pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >:

Public Types
using	ssv = Eigen::Matrix< float_t, dimstate, 1 >

using	osv = Eigen::Matrix< float_t, dimobs, 1 >

using	isv = Eigen::Matrix< float_t, diminput, 1 >

using	ssMat = Eigen::Matrix< float_t, dimstate, dimstate >

using	osMat = Eigen::Matrix< float_t, dimobs, dimobs >

using	siMat = Eigen::Matrix< float_t, dimstate, diminput >

using	oiMat = Eigen::Matrix< float_t, dimobs, diminput >

using	obsStateSizeMat = Eigen::Matrix< float_t, dimobs, dimstate >

using	stateObsSizeMat = Eigen::Matrix< float_t, dimstate, dimobs >

Public Types inherited from pf::bases::cf_filter< dimstate, dimobs, float_t >
using	obs_sized_vec = Eigen::Matrix< float_t, dimobs, 1 >

using	state_sized_vec = Eigen::Matrix< float_t, dimstate, 1 >

Public Member Functions
	kalman ()
	Default constructor. More...

	kalman (const ssv &initStateMean, const ssMat &initStateVar)
	Non-default constructor. More...

virtual	~kalman ()
	The (virtual) destructor.

float_t	getLogCondLike () const
	returns the log of the latest conditional likelihood. More...

ssv	getFiltMean () const
	Get the current filter mean. More...

ssMat	getFiltVar () const
	Get the current filter variance-covariance matrix. More...

osv	getPredYMean (const ssMat &stateTrans, const obsStateSizeMat &obsMat, const siMat &stateInptAffector, const oiMat &obsInptAffector, const isv &inputData) const
	get the one-step-ahead point forecast for y More...

osMat	getPredYVar (const ssMat &stateTrans, const ssMat &cholStateVar, const obsStateSizeMat &obsMat, const osMat &cholObsVar) const
	get the one-step-ahead forecast variance More...

void	update (const osv &yt, const ssMat &stateTrans, const ssMat &cholStateVar, const siMat &stateInptAffector, const isv &inputData, const obsStateSizeMat &obsMat, const oiMat &obsInptAffector, const osMat &cholObsVar)
	Perform a Kalman filter predict-and-update. More...

Public Member Functions inherited from pf::bases::cf_filter< dimstate, dimobs, float_t >
virtual	~cf_filter ()
	The (virtual) destructor.

Private Member Functions
void	updatePrior (const ssMat &stateTransMat, const ssMat &cholStateVar, const siMat &stateInptAffector, const isv &inputData)
	Predicts the next state. More...

void	updatePosterior (const osv &yt, const obsStateSizeMat &obsMat, const oiMat &obsInptAffector, const isv &inputData, const osMat &cholObsVar)
	Turns prediction into new filtering distribution. More...

Private Attributes
ssv	m_predMean
	predictive state mean

ssv	m_filtMean
	filter mean

ssMat	m_predVar
	predictive var matrix

ssMat	m_filtVar
	filter var matrix

float_t	m_lastLogCondLike
	latest log conditional likelihood

bool	m_fresh
	has data been observed?

const float_t	m_pi
	pi

Detailed Description

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t, bool debug = false>
class pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >

A class template for Kalman filtering.

Author: taylor

Member Typedef Documentation

◆ isv

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug = false>

using pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::isv = Eigen::Matrix<float_t,diminput,1>

"input size vector" type alias for linear algebra stuff

◆ obsStateSizeMat

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug = false>

using pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::obsStateSizeMat = Eigen::Matrix<float_t,dimobs,dimstate>

"observation dimension by state dimension -sized matrix"

◆ oiMat

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug = false>

using pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::oiMat = Eigen::Matrix<float_t,dimobs,diminput>

"observation dimension by input dim matrix"

◆ osMat

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug = false>

using pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::osMat = Eigen::Matrix<float_t,dimobs,dimobs>

"observation size matrix" type alias for linear algebra stuff

◆ osv

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug = false>

using pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::osv = Eigen::Matrix<float_t,dimobs,1>

"observation size vector" type alias for linear algebra stuff

◆ siMat

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug = false>

using pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::siMat = Eigen::Matrix<float_t,dimstate,diminput>

"state dim by input dimension matrix"

◆ ssMat

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug = false>

using pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::ssMat = Eigen::Matrix<float_t,dimstate,dimstate>

"state size matrix" type alias for linear algebra stuff

◆ ssv

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug = false>

using pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::ssv = Eigen::Matrix<float_t,dimstate,1>

"state size vector" type alias for linear algebra stuff

◆ stateObsSizeMat

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug = false>

using pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::stateObsSizeMat = Eigen::Matrix<float_t,dimstate,dimobs>

"state dimension by observation dimension matrix

Constructor & Destructor Documentation

◆ kalman() [1/2]

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug>

pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::kalman

Default constructor.

Need ths fir constructing default std::array<>s. Fills all vectors and matrices with zeros.

◆ kalman() [2/2]

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug>

pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::kalman	(	const ssv &	initStateMean,
		const ssMat &	initStateVar
	)

Non-default constructor.

Member Function Documentation

◆ getFiltMean()

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug>

auto pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::getFiltMean

Get the current filter mean.

Returns: E[x_t | y_{1:t}]

◆ getFiltVar()

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug>

auto pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::getFiltVar

Get the current filter variance-covariance matrix.

Returns: V[x_t | y_{1:t}]

◆ getLogCondLike()

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug>

float_t pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::getLogCondLike

virtual

returns the log of the latest conditional likelihood.

Returns: log p(y_t | y_{1:t-1}) or log p(y_1)

Implements pf::bases::cf_filter< dimstate, dimobs, float_t >.

◆ getPredYMean()

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug>

auto pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::getPredYMean	(	const ssMat &	stateTrans,
		const obsStateSizeMat &	obsMat,
		const siMat &	stateInptAffector,
		const oiMat &	obsInptAffector,
		const isv &	inputData
	)		const

get the one-step-ahead point forecast for y

Returns: E[y_{t+1} | y_{1:t}, params]

◆ getPredYVar()

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug>

auto pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::getPredYVar	(	const ssMat &	stateTrans,
		const ssMat &	cholStateVar,
		const obsStateSizeMat &	obsMat,
		const osMat &	cholObsVar
	)		const

get the one-step-ahead forecast variance

Returns: V[y_{t+1} | y_{1:t}, params]

◆ update()

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug>

void pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::update	(	const osv &	yt,
		const ssMat &	stateTrans,
		const ssMat &	cholStateVar,
		const siMat &	stateInptAffector,
		const isv &	inputData,
		const obsStateSizeMat &	obsMat,
		const oiMat &	obsInptAffector,
		const osMat &	cholObsVar
	)

Perform a Kalman filter predict-and-update.

Parameters

yt	the new data point.
stateTrans	the transition matrix of the state
cholStateVar	the Cholesky Decomposition of the state noise covariance matrix.
stateInptAffector	the matrix affecting how input data affects state transition.
inputData	exogenous input data
obsMat	the observation/emission matrix of the observation's conditional (on the state) distn.
obsInptAffector	the matrix affecting how input data affects the observational distribution.
cholObsVar	the Cholesky Decomposition of the observatio noise covariance matrix.

◆ updatePosterior()

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug>

void pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::updatePosterior	(	const osv &	yt,
		const obsStateSizeMat &	obsMat,
		const oiMat &	obsInptAffector,
		const isv &	inputData,
		const osMat &	cholObsVar
	)

private

Turns prediction into new filtering distribution.

Parameters

yt
obsMat
obsInptAffector
inputData
cholObsVar

◆ updatePrior()

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t , bool debug>

void pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >::updatePrior	(	const ssMat &	stateTransMat,
		const ssMat &	cholStateVar,
		const siMat &	stateInptAffector,
		const isv &	inputData
	)

private

Predicts the next state.

Todo:: handle diagonal variance matrices, and ensure symmetricness in other ways

Parameters

stateTransMat
cholStateVar
stateInptAffector
inputData

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

include/pf/cf_filters.h

Public Types

Public Member Functions

Private Member Functions

Private Attributes

Detailed Description

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t, bool debug = false> class pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >

Member Typedef Documentation

◆ isv

◆ obsStateSizeMat

◆ oiMat

◆ osMat

◆ osv

◆ siMat

◆ ssMat

◆ ssv

◆ stateObsSizeMat

Constructor & Destructor Documentation

◆ kalman() [1/2]

◆ kalman() [2/2]

Member Function Documentation

◆ getFiltMean()

◆ getFiltVar()

◆ getLogCondLike()

◆ getPredYMean()

◆ getPredYVar()

◆ update()

◆ updatePosterior()

◆ updatePrior()

template<size_t dimstate, size_t dimobs, size_t diminput, typename float_t, bool debug = false>
class pf::filters::kalman< dimstate, dimobs, diminput, float_t, debug >