Unscented Kalman Filter¶

Sequential state estimator using sigma points to capture nonlinear dynamics without linearization.

UnscentedKalmanFilter ¶

UnscentedKalmanFilter(epoch: Any, state: Any, initial_covariance: Any, propagation_config: Any, force_config: Any, measurement_models: Any, config: Any = None, params: Any = None, additional_dynamics: Any = None, control_input: Any = None)

Unscented Kalman Filter for sequential state estimation.

Uses sigma points to propagate state statistics through nonlinear dynamics and measurement models without linearization. Does not require Jacobians or STM propagation.

Example

import brahe as bh
import numpy as np

bh.initialize_eop()

epoch = bh.Epoch(2024, 1, 1, 0, 0, 0.0)
state = np.array([6878e3, 0.0, 0.0, 0.0, 7612.0, 0.0])
p0 = np.diag([1e6, 1e6, 1e6, 1e2, 1e2, 1e2])

ukf = bh.UnscentedKalmanFilter(
    epoch, state, p0,
    propagation_config=bh.NumericalPropagationConfig.default(),
    force_config=bh.ForceModelConfig.two_body(),
    measurement_models=[bh.InertialPositionMeasurementModel(10.0)],
)

Initialize instance.

current_covariance `method descriptor` ¶

current_covariance() -> ndarray

Get current covariance estimate.

Returns:

Type	Description
`ndarray`	numpy.ndarray: Current covariance matrix.

current_epoch `method descriptor` ¶

current_epoch() -> Epoch

Get current epoch.

Returns:

Name	Type	Description
`Epoch`	`Epoch`	Current filter epoch.

current_state `method descriptor` ¶

current_state() -> ndarray

Get current state estimate.

Returns:

Type	Description
`ndarray`	numpy.ndarray: Current state vector.

process_observation `method descriptor` ¶

process_observation(observation: Observation) -> FilterRecord

Process a single observation.

Parameters:

Name	Type	Description	Default
`observation`	`Observation`	The observation to process.	required

Returns:

Name	Type	Description
`FilterRecord`	`FilterRecord`	Record containing pre/post-fit residuals, gain, etc.

process_observations `method descriptor` ¶

process_observations(observations: list[Observation]) -> Any

Process multiple observations (auto-sorted by epoch).

Parameters:

Name	Type	Description	Default
`observations`	`list[Observation]`	List of observations.	required

records `method descriptor` ¶

records() -> list[FilterRecord]

Get all stored filter records.

Returns:

Type	Description
`list[FilterRecord]`	list[FilterRecord]: List of filter records.

UKFConfig ¶

UKFConfig(state_dim: Any = 6, alpha: Any = 0.001, beta: Any = 2.0, kappa: Any = 0.0, process_noise: Any = None, store_records: Any = True)

Configuration for the Unscented Kalman Filter.

Example

import brahe as bh
config = bh.UKFConfig(state_dim=6, alpha=1e-3, beta=2.0, kappa=0.0)

Initialize instance.

alpha `property` ¶

alpha: Any

beta `property` ¶

beta: Any

kappa `property` ¶

kappa: Any

state_dim `property` ¶

state_dim: ndarray

store_records `property` ¶

store_records: Any

default `staticmethod` ¶

default() -> UKFConfig

Create a default UKF configuration.

Returns:

Name	Type	Description
`UKFConfig`	`UKFConfig`	Default configuration (state_dim=6, alpha=1e-3, beta=2.0, kappa=0.0).

Unscented Kalman Filter¶