Package org.hipparchus.filtering.kalman.extended

Class ExtendedKalmanFilter<T extends Measurement>

java.lang.Object

org.hipparchus.filtering.kalman.AbstractKalmanFilter<T>

org.hipparchus.filtering.kalman.extended.ExtendedKalmanFilter<T>

Type Parameters:

T - the type of the measurements

All Implemented Interfaces:

KalmanEstimate, KalmanFilter<T>

public class ExtendedKalmanFilter<T extends Measurement>
extends AbstractKalmanFilter<T>

Kalman filter for non-linear process.

Since:

1.3

Constructor Summary

Constructors

Constructor	Description
ExtendedKalmanFilter(MatrixDecomposer decomposer, NonLinearProcess<T> process, ProcessEstimate initialState)	Simple constructor.

Method Summary

Modifier and Type	Method	Description
ProcessEstimate	estimationStep(T measurement)	Perform one estimation step.

Methods inherited from class org.hipparchus.filtering.kalman.AbstractKalmanFilter

computeInnovationCovarianceMatrix, correct, getCorrected, getObserver, getPredicted, getStateCrossCovariance, predict, setObserver

Methods inherited from class java.lang.Object

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

Constructor Detail

ExtendedKalmanFilter

public ExtendedKalmanFilter(MatrixDecomposer decomposer,
                            NonLinearProcess<T> process,
                            ProcessEstimate initialState)

Simple constructor.

Parameters:

decomposer - decomposer to use for the correction phase

process - non-linear process to estimate

initialState - initial state

Method Detail

estimationStep

public ProcessEstimate estimationStep(T measurement)
                               throws MathRuntimeException

Perform one estimation step.

Parameters:

measurement - single measurement to handle

Returns:

estimated state after measurement has been considered

Throws:

MathRuntimeException - if estimation fails