org.hipparchus.filtering.kalman

Class ProcessEstimate

java.lang.Object

org.hipparchus.filtering.kalman.ProcessEstimate

public class ProcessEstimate
extends Object

Holder for process state and covariance.

The estimate always contains time, state and covariance. These data are the only ones needed to start a Kalman filter. Once a filter has been started and produces new estimates, these new estimates will always contain a state transition matrix and if the measurement has not been ignored, they will also contain measurement Jacobian, innovation covariance and Kalman gain.

Since:

1.3

Constructor Summary

Constructors

Constructor and Description
ProcessEstimate(double time, RealVector state, RealMatrix covariance) Simple constructor.
ProcessEstimate(double time, RealVector state, RealMatrix covariance, RealMatrix stateTransitionMatrix, RealMatrix measurementJacobian, RealMatrix innovationCovariance, RealMatrix kalmanGain) Simple constructor.

Method Summary

Modifier and Type	Method and Description
RealMatrix	getCovariance() Get the state covariance.
RealMatrix	getInnovationCovariance() Get the innovation covariance matrix.
RealMatrix	getKalmanGain() Get the Kalman gain matrix.
RealMatrix	getMeasurementJacobian() Get the Jacobian of the measurement with respect to the state (H matrix).
RealVector	getState() Get the state vector.
RealMatrix	getStateTransitionMatrix() Get state transition matrix between previous state and estimated (but not yet corrected) state.
double	getTime() Get the process time.

Methods inherited from class java.lang.Object

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

Constructor Detail

ProcessEstimate

public ProcessEstimate(double time,
                       RealVector state,
                       RealMatrix covariance)

Simple constructor.

This constructor sets state transition matrix, covariance matrix H, innovation covariance matrix and Kalman gain k to null.

Parameters:

time - process time (typically the time or index of a measurement)

state - state vector

covariance - state covariance

ProcessEstimate

public ProcessEstimate(double time,
                       RealVector state,
                       RealMatrix covariance,
                       RealMatrix stateTransitionMatrix,
                       RealMatrix measurementJacobian,
                       RealMatrix innovationCovariance,
                       RealMatrix kalmanGain)

Simple constructor.

Parameters:

time - process time (typically the time or index of a measurement)

state - state vector

covariance - state covariance

stateTransitionMatrix - state transition matrix between previous state and estimated (but not yet corrected) state

measurementJacobian - Jacobian of the measurement with respect to the state

innovationCovariance - innovation covariance matrix, defined as \(h.P.h^T + r\), may be null

kalmanGain - Kalman Gain matrix, may be null

Since:

1.4

Method Detail

getTime

public double getTime()

Get the process time.

Returns:

process time (typically the time or index of a measurement)

getState

public RealVector getState()

Get the state vector.

Returns:

state vector

getCovariance

public RealMatrix getCovariance()

Get the state covariance.

Returns:

state covariance

getStateTransitionMatrix

public RealMatrix getStateTransitionMatrix()

Get state transition matrix between previous state and estimated (but not yet corrected) state.

Returns:

state transition matrix between previous state and estimated state (but not yet corrected) (may be null for initial process estimate)

Since:

1.4

getMeasurementJacobian

public RealMatrix getMeasurementJacobian()

Get the Jacobian of the measurement with respect to the state (H matrix).

Returns:

Jacobian of the measurement with respect to the state (may be null for initial process estimate or if the measurement has been ignored)

Since:

1.4

getInnovationCovariance

public RealMatrix getInnovationCovariance()

Get the innovation covariance matrix.

Returns:

innovation covariance matrix (may be null for initial process estimate or if the measurement has been ignored)

Since:

1.4

getKalmanGain

public RealMatrix getKalmanGain()

Get the Kalman gain matrix.

Returns:

Kalman gain matrix (may be null for initial process estimate or if the measurement has been ignored)

Since:

1.4