Smoothers¶

class stonesoup.smoother.base.Smoother(transition_model=None)[source]¶

Smoother Base Class.

Parameters: transition_model (TransitionModel, optional) – Transition Model.

transition_model: stonesoup.models.transition.base.TransitionModel¶: Transition Model.

Linear Gaussian¶

class stonesoup.smoother.lineargaussian.Backward(transition_model=None)[source]¶

Backwards component of a fixed-interval forward-backward smoother for a Linear Gaussian State Space Model.

Parameters: transition_model (TransitionModel, optional) – Transition Model.

track_smooth(filtered_track)[source]¶

Apply smoothing to a track of filtered estimates.

Parameters: filtered_track (Track) – Track object consisting a of GaussianStateUpdate objects.
Returns: smoothed_track – Track object containing smoothed GaussianState objects.
Return type: Track

smooth(filtered_state_t, predicted_state_tplus1, smoothed_state_tplus1)[source]¶

Deduce smoothed distribution \(p(x_{t} | y_{1:T})\) from \(p(x_{t} | y_{1:t})\), \(p(x_{t+1} | y_{1:t})\) and \(p(x_{t+1} | y_{1:T})\).

Parameters

filtered_state_t (GaussianState) – Filtered state at time t.
predicted_state_tplus1 (GaussianState) – Prediction (at timestep t), of the state at time t+1.
smoothed_state_tplus1 (GaussianState) – Smoothed state at time t+1.

Returns

Return type

GaussianState