org.orekit.estimation.sequential

Class UnscentedKalmanEstimatorBuilder

java.lang.Object

org.orekit.estimation.sequential.UnscentedKalmanEstimatorBuilder

public class UnscentedKalmanEstimatorBuilder
extends Object

Builder for an Unscented Kalman filter estimator.

The builder is generalized to accept any PropagatorBuilder. Howerver, it is absolutely not recommended to use a DSSTPropagatorBuilder. A specific semi-analytical unscented Kalman Filter is implemented and shall be used.

Since:

11.3

Author:

Gaëtan Pierre, Bryan Cazabonne

Constructor Summary

Constructors

Constructor and Description
UnscentedKalmanEstimatorBuilder() Default constructor.

Method Summary

Modifier and Type	Method and Description
UnscentedKalmanEstimatorBuilder	addPropagationConfiguration(PropagatorBuilder builder, CovarianceMatrixProvider provider) Add a propagation configuration.
UnscentedKalmanEstimator	build() Construct a UnscentedKalmanEstimator from the data in this builder.
UnscentedKalmanEstimatorBuilder	decomposer(org.hipparchus.linear.MatrixDecomposer matrixDecomposer) Configure the matrix decomposer.
UnscentedKalmanEstimatorBuilder	estimatedMeasurementsParameters(ParameterDriversList estimatedMeasurementsParams, CovarianceMatrixProvider provider) Configure the estimated measurement parameters.
UnscentedKalmanEstimatorBuilder	unscentedTransformProvider(org.hipparchus.util.UnscentedTransformProvider transformProvider) Configure the unscented transform provider.

Methods inherited from class java.lang.Object

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

Constructor Detail

UnscentedKalmanEstimatorBuilder

public UnscentedKalmanEstimatorBuilder()

Default constructor. Set an Unscented Kalman filter.

Method Detail

build

public UnscentedKalmanEstimator build()

Construct a UnscentedKalmanEstimator from the data in this builder.

Before this method is called, addPropagationConfiguration() must have been called at least once, otherwise configuration is incomplete and an exception will be raised.

In addition, the unscentedTransformProvider() must be called to configure the unscented transform provider use during the estimation process, otherwise configuration is incomplete and an exception will be raised.

Returns:

a new UnscentedKalmanEstimator.

decomposer

public UnscentedKalmanEstimatorBuilder decomposer(org.hipparchus.linear.MatrixDecomposer matrixDecomposer)

Configure the matrix decomposer.

Parameters:

matrixDecomposer - decomposer to use for the correction phase

Returns:

this object.

unscentedTransformProvider

public UnscentedKalmanEstimatorBuilder unscentedTransformProvider(org.hipparchus.util.UnscentedTransformProvider transformProvider)

Configure the unscented transform provider.

Parameters:

transformProvider - unscented transform to use for the prediction phase

Returns:

this object.

addPropagationConfiguration

public UnscentedKalmanEstimatorBuilder addPropagationConfiguration(PropagatorBuilder builder,
                                                                   CovarianceMatrixProvider provider)

Add a propagation configuration.

This method must be called once for each propagator to managed with the unscented kalman estimatior. The propagators order in the Kalman filter will be the call order.

The provider should return a matrix with dimensions and ordering consistent with the builder configuration. The first 6 rows/columns correspond to the 6 orbital parameters which must all be present, regardless of the fact they are estimated or not. The remaining elements correspond to the subset of propagation parameters that are estimated, in the same order as propagatorBuilder.getPropagationParametersDrivers().getDrivers() (but filtering out the non selected drivers).

Parameters:

builder - The propagator builder to use in the Kalman filter.

provider - The process noise matrices provider to use, consistent with the builder.

Returns:

this object.

See Also:

getProcessNoiseMatrix(previous, current)

estimatedMeasurementsParameters

public UnscentedKalmanEstimatorBuilder estimatedMeasurementsParameters(ParameterDriversList estimatedMeasurementsParams,
                                                                       CovarianceMatrixProvider provider)

Configure the estimated measurement parameters.

If this method is not called, no measurement parameters will be estimated.

Parameters:

estimatedMeasurementsParams - The estimated measurements' parameters list.

provider - covariance matrix provider for the estimated measurement parameters

Returns:

this object.