Package org.hipparchus.util

Class MerweUnscentedTransform

java.lang.Object

org.hipparchus.util.AbstractUnscentedTransform

org.hipparchus.util.MerweUnscentedTransform

All Implemented Interfaces:

UnscentedTransformProvider

public class MerweUnscentedTransform
extends AbstractUnscentedTransform

Unscented transform as defined by Merwe and Wan.

The unscented transform uses three parameters: alpha, beta and kappa. Alpha determines the spread of the sigma points around the process state, kappa is a secondary scaling parameter, and beta is used to incorporate prior knowledge of the distribution of the process state.

Since:

2.2

See Also:

• "E. A. Wan and R. Van der Merwe, The unscented Kalman filter for nonlinear estimation, in Proc. Symp. Adaptive Syst. Signal Process., Commun. Contr., Lake Louise, AB, Canada, Oct. 2000."

Field Summary

Fields

Modifier and Type	Field	Description
static final double	DEFAULT_ALPHA	Default value for alpha (0.5, see reference).
static final double	DEFAULT_BETA	Default value for beta (2.0, see reference).
static final double	DEFAULT_KAPPA	Default value for kappa, (0.0, see reference).

Constructor Summary

Constructors

Constructor	Description
MerweUnscentedTransform(int stateDim)	Default constructor.
MerweUnscentedTransform(int stateDim, double alpha, double beta, double kappa)	Simple constructor.

Method Summary

Modifier and Type	Method	Description
protected double	getMultiplicationFactor()	Get the factor applied to the covariance matrix during the unscented transform.
RealVector	getWc()	Get the covariance weights.
RealVector	getWm()	Get the mean weights.

Methods inherited from class org.hipparchus.util.AbstractUnscentedTransform

unscentedTransform

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface org.hipparchus.util.UnscentedTransformProvider

getUnscentedCovariance, getUnscentedMeanState, inverseUnscentedTransform

Field Details

DEFAULT_ALPHA

public static final double DEFAULT_ALPHA

Default value for alpha (0.5, see reference).

See Also:

• Constant Field Values

DEFAULT_BETA

public static final double DEFAULT_BETA

Default value for beta (2.0, see reference).

See Also:

• Constant Field Values

DEFAULT_KAPPA

public static final double DEFAULT_KAPPA

Default value for kappa, (0.0, see reference).

See Also:

• Constant Field Values

Constructor Details

MerweUnscentedTransform

public MerweUnscentedTransform(int stateDim)

Default constructor.

This constructor uses default values for alpha, beta, and kappa.

Parameters:

stateDim - the dimension of the state

See Also:

• DEFAULT_ALPHA
• DEFAULT_BETA
• DEFAULT_KAPPA
• MerweUnscentedTransform(int, double, double, double)

MerweUnscentedTransform

public MerweUnscentedTransform(int stateDim,
 double alpha,
 double beta,
 double kappa)

Simple constructor.

Parameters:

stateDim - the dimension of the state

alpha - scaling control parameter (determines the spread of the sigma points around the process state)

beta - free parameter (used to incorporate prior knowledge of the distribution of the process state)

kappa - secondary scaling factor (usually set to 0.0)

Method Details

getWc

public RealVector getWc()

Get the covariance weights.

Returns:

the covariance weights

getWm

public RealVector getWm()

Get the mean weights.

Returns:

the mean weights

getMultiplicationFactor

protected double getMultiplicationFactor()

Get the factor applied to the covariance matrix during the unscented transform.

Specified by:

getMultiplicationFactor in class AbstractUnscentedTransform

Returns:

the factor applied to the covariance matrix during the unscented transform