org.orekit.propagation.conversion

Class AbstractPropagatorConverter

java.lang.Object

org.orekit.propagation.conversion.AbstractPropagatorConverter

All Implemented Interfaces:

PropagatorConverter

Direct Known Subclasses:

FiniteDifferencePropagatorConverter, JacobianPropagatorConverter

public abstract class AbstractPropagatorConverter
extends Object
implements PropagatorConverter

Common handling of PropagatorConverter methods for propagators conversions.

This abstract class factors the common code for propagators conversion. Only one method must be implemented by derived classes: getObjectiveFunction().

The converter uses the LevenbergMarquardtOptimizer from the Hipparchus library. Different implementations correspond to different methods for computing the Jacobian.

Since:

6.0

Author:

Pascal Parraud

Constructor Summary

Constructors

Modifier	Constructor and Description
protected	AbstractPropagatorConverter(PropagatorBuilder builder, double threshold, int maxIterations) Build a new instance.

Method Summary

Modifier and Type	Method and Description
Propagator	convert(List<SpacecraftState> states, boolean positionOnly, List<String> freeParameters) Find the propagator that minimize the mean square error for a sample of states.
Propagator	convert(List<SpacecraftState> states, boolean positionOnly, String... freeParameters) Find the propagator that minimize the mean square error for a sample of states.
Propagator	convert(Propagator source, double timeSpan, int nbPoints, List<String> freeParameters) Convert a propagator to another.
Propagator	convert(Propagator source, double timeSpan, int nbPoints, String... freeParameters) Convert a propagator to another.
Propagator	getAdaptedPropagator() Get the adapted propagator.
int	getEvaluations() Get the number of objective function evaluations.
protected Frame	getFrame() Get the frame of the initial state.
protected abstract org.hipparchus.optim.nonlinear.vector.leastsquares.MultivariateJacobianFunction	getModel() Get the Jacobian of the function computing position/velocity at sample points.
protected abstract org.hipparchus.analysis.MultivariateVectorFunction	getObjectiveFunction() Get the function computing position/velocity at sample points.
double	getRMS() Get the Root Mean Square Deviation of the fitting.
protected List<SpacecraftState>	getSample() Get the states sample.
protected int	getTargetSize() Get the size of the target.
protected boolean	isOnlyPosition() Check if fitting uses only sample positions.

Methods inherited from class java.lang.Object

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

Constructor Detail

AbstractPropagatorConverter

protected AbstractPropagatorConverter(PropagatorBuilder builder,
                                      double threshold,
                                      int maxIterations)

Build a new instance.

Parameters:

builder - propagator builder

threshold - absolute convergence threshold for optimization algorithm

maxIterations - maximum number of iterations for fitting

Method Detail

convert

public Propagator convert(Propagator source,
                          double timeSpan,
                          int nbPoints,
                          List<String> freeParameters)
                   throws OrekitException,
                          IllegalArgumentException

Convert a propagator to another.

Specified by:

convert in interface PropagatorConverter

Parameters:

source - initial propagator (the propagator will be used for sample generation, if it is a numerical propagator, its initial state will be reset unless AbstractIntegratedPropagator.setResetAtEnd(boolean) has been called beforehand)

timeSpan - time span for fitting

nbPoints - number of fitting points over time span

freeParameters - names of the free parameters

Returns:

adapted propagator

Throws:

OrekitException - if propagator cannot be adapted

IllegalArgumentException - if one of the parameters cannot be free

convert

public Propagator convert(Propagator source,
                          double timeSpan,
                          int nbPoints,
                          String... freeParameters)
                   throws OrekitException,
                          IllegalArgumentException

Convert a propagator to another.

Specified by:

convert in interface PropagatorConverter

Parameters:

source - initial propagator (the propagator will be used for sample generation, if it is a numerical propagator, its initial state will be reset unless AbstractIntegratedPropagator.setResetAtEnd(boolean) has been called beforehand)

timeSpan - time span for fitting

nbPoints - number of fitting points over time span

freeParameters - names of the free parameters

Returns:

adapted propagator

Throws:

OrekitException - if propagator cannot be adapted

IllegalArgumentException - if one of the parameters cannot be free

convert

public Propagator convert(List<SpacecraftState> states,
                          boolean positionOnly,
                          List<String> freeParameters)
                   throws OrekitException,
                          IllegalArgumentException

Find the propagator that minimize the mean square error for a sample of states.

Specified by:

convert in interface PropagatorConverter

Parameters:

states - spacecraft states sample to fit

positionOnly - if true, consider only position data otherwise both position and velocity are used

freeParameters - names of the free parameters

Returns:

adapted propagator

Throws:

OrekitException - if propagator cannot be adapted

IllegalArgumentException - if one of the parameters cannot be free

convert

public Propagator convert(List<SpacecraftState> states,
                          boolean positionOnly,
                          String... freeParameters)
                   throws OrekitException,
                          IllegalArgumentException

Find the propagator that minimize the mean square error for a sample of states.

Specified by:

convert in interface PropagatorConverter

Parameters:

states - spacecraft states sample to fit

positionOnly - if true, consider only position data otherwise both position and velocity are used

freeParameters - names of the free parameters

Returns:

adapted propagator

Throws:

OrekitException - if propagator cannot be adapted

IllegalArgumentException - if one of the parameters cannot be free

getAdaptedPropagator

public Propagator getAdaptedPropagator()

Get the adapted propagator.

Returns:

adapted propagator

getRMS

public double getRMS()

Get the Root Mean Square Deviation of the fitting.

Returns:

RMSD

getEvaluations

public int getEvaluations()

Get the number of objective function evaluations.

Returns:

the number of objective function evaluations.

getObjectiveFunction

protected abstract org.hipparchus.analysis.MultivariateVectorFunction getObjectiveFunction()

Get the function computing position/velocity at sample points.

Returns:

function computing position/velocity at sample points

getModel

protected abstract org.hipparchus.optim.nonlinear.vector.leastsquares.MultivariateJacobianFunction getModel()

Get the Jacobian of the function computing position/velocity at sample points.

Returns:

Jacobian of the function computing position/velocity at sample points

isOnlyPosition

protected boolean isOnlyPosition()

Check if fitting uses only sample positions.

Returns:

true if fitting uses only sample positions

getTargetSize

protected int getTargetSize()

Get the size of the target.

Returns:

target size

getFrame

protected Frame getFrame()

Get the frame of the initial state.

Returns:

the orbit frame

getSample

protected List<SpacecraftState> getSample()

Get the states sample.

Returns:

the states sample