PARP Research Group Universidad de Murcia


QVUKF Class Reference
[Signal processing functionality]

Implementation of UKF filter. More...

#include <qvukf.h>

List of all members.

Public Member Functions

 QVUKF (double k=0.0, double alpha=0.5, double beta=2.0)
 QVUKF default constructor.
 QVUKF (const QVUKFstate &state)
 QVUKF copy constructor.
void setState (const QVUKFstate &state)
 Method that fills the current state from a given QVUKFstate.
void update (const QVMatrix &obs, QVFunction< QVVector, QVVector > &g, QVFunction< QVVector, QVVector > &h, const QVMatrix &Rt, const QVMatrix &Qt)
 Method that updates the current state, moving from X_k to X_k+1.
QVMatrix getState () const
 Method that returns the current system state.
double getDiscrepance () const
 Method that returns the discrepancy.

Detailed Description

Implementation of UKF filter.

This is an implementation of the general Unscented Kalman Filter, useful for non-linear filtering.

Definition at line 73 of file qvukf.h.


Constructor & Destructor Documentation

QVUKF::QVUKF ( double  k = 0.0,
double  alpha = 0.5,
double  beta = 2.0 
)

QVUKF default constructor.

Parameters:
k. 
alpha. 
beta. 
QVUKF::QVUKF ( const QVUKFstate state  ) 

QVUKF copy constructor.

Parameters:
state A QVUKFstate.

Definition at line 26 of file qvukf.cpp.


Member Function Documentation

void QVUKF::setState ( const QVUKFstate state  ) 

Method that fills the current state from a given QVUKFstate.

Parameters:
state QVUKFstate used to fill the current filter state (discrepancy is reset to 0).

Definition at line 38 of file qvukf.cpp.

void QVUKF::update ( const QVMatrix obs,
QVFunction< QVVector, QVVector > &  g,
QVFunction< QVVector, QVVector > &  h,
const QVMatrix Rt,
const QVMatrix Qt 
)

Method that updates the current state, moving from X_k to X_k+1.

Parameters:
obs QVMatrix that contains an acquired observation.
g A functor that implements the dynamic model for a specific problem.
h A functor that implements how to propagate the observation according to the conditions of a specific problem.
Rt Multivariate Gaussian noise (0-mean) that will be added to the covariance matrix.
Qt Multivariate Gaussian noise (0-mean) that will be added to the predicted covariance matrix.

Definition at line 160 of file qvukf.cpp.

QVMatrix QVUKF::getState (  )  const

Method that returns the current system state.

Returns:
A QVMatrix that contains the current system state.

Definition at line 45 of file qvukf.cpp.

double QVUKF::getDiscrepance (  )  const

Method that returns the discrepancy.

Returns:
The system discrepancy.

Definition at line 50 of file qvukf.cpp.


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



QVision framework. PARP research group. Copyright © 2007, 2008, 2009, 2010, 2011.