ADCS.estimators.estimator_helpers.estimator_helpers module

class ADCS.estimators.estimator_helpers.estimator_helpers.EstimatedArray(val, cov=<factory>, int_cov=<factory>)

Bases: object

Represents an estimated vector with associated covariance and integrated covariance.

This structure is commonly used for parameter estimation where the state is a generic vector rather than a specific orbital state.

Parameters:

• val (np.ndarray) – The estimated state or parameter vector \(\hat{\mathbf{x}}\).

• cov (np.ndarray) – The covariance matrix \(P\) representing uncertainty in val. Must be square and match the dimensions of val.

• int_cov (np.ndarray) – The integrated process covariance \(Q_{int}\) (e.g., accumulated process noise over time).

copy()

Creates a deep copy of the EstimatedArray.

Returns:

A new instance with independent data copies.

Return type:

EstimatedArray

pull_indices(inds_mask, cov_missing_inds=None)

Extracts a sub-estimate and associated covariance blocks based on a mask.

This is useful for isolating specific parameters from a larger state vector.

Parameters:

• inds_mask (list or np.ndarray) – Indices of the values to extract.

• cov_missing_inds (list or np.ndarray, optional) – Indices to exclude from the covariance extraction (optional). If None, inds_mask is used for covariance as well.

Returns:

A new EstimatedArray containing only the selected elements.

Return type:

EstimatedArray

set_indices(inds_mask, val, cov, int_cov, cov_missing_inds=None)

Inserts values and covariance blocks back into this estimate.

Parameters:

• inds_mask (list or np.ndarray) – Indices in the full vector where data should be inserted.

• val (np.ndarray) – The sub-vector of values to insert.

• cov (np.ndarray) – The covariance block corresponding to the inserted values.

• int_cov (np.ndarray) – The integrated covariance block to insert.

• cov_missing_inds (list or np.ndarray, optional) – Indices to exclude from the covariance insertion mapping (optional).

cov: ndarray

int_cov: ndarray

val: ndarray

class ADCS.estimators.estimator_helpers.estimator_helpers.EstimatedOrbital_State(os, P=<factory>, Q=<factory>)

Bases: object

A container for an Orbital State and its associated estimation uncertainty.

This class wraps the physical Orbital_State with estimation statistics, specifically the state covariance \(P\) and process noise \(Q\).

Parameters:

• os (Orbital_State) – The estimated physical state (Position \(\mathbf{r}\) and Velocity \(\mathbf{v}\)).

• P (np.ndarray) – The \(6 \times 6\) state covariance matrix. .. math:: P = E[(hat{x} - x)(hat{x} - x)^T]

• Q (np.ndarray) – The \(6 \times 6\) process noise covariance matrix.

copy()

Creates a deep copy of the EstimatedOrbital_State.

Returns:

A new instance with deep-copied state and matrices.

Return type:

EstimatedOrbital_State

pull_indices(inds_mask, cov_missing_inds=None)

Extracts a partial state estimate based on indices mapping to [R, V].

The mapping is linear: indices 0-2 correspond to Position (R), and 3-5 to Velocity (V).

Parameters:

• inds_mask (list or np.ndarray) – The indices of the 6-element state vector to extract.

• cov_missing_inds (list or np.ndarray, optional) – Indices to exclude from covariance slicing (optional).

Returns:

A new EstimatedOrbital_State containing the subset of state and covariance. Note: The underlying Orbital_State will have 0.0 in unselected components.

Return type:

EstimatedOrbital_State

set_indices(inds_mask, val, P, Q, cov_missing_inds=None)

Inserts values and covariance blocks back into the full 6-state estimate.

Parameters:

• inds_mask (list or np.ndarray) – Indices in the full 6-element vector where data should be inserted.

• val (np.ndarray) – The values to insert into Position/Velocity.

• P (np.ndarray) – The covariance block \(P_{sub}\) to insert.

• Q (np.ndarray) – The process noise block \(Q_{sub}\) to insert.

• cov_missing_inds (list or np.ndarray, optional) – Indices to exclude from covariance insertion (optional).

P: ndarray

Q: ndarray

os: Orbital_State