states.KalmanCentroid2D
class KalmanCentroid2D:Constant-velocity 2-D centroid Kalman process.
State is 4-D ``[x, y, vx, vy]``; measurement is 2-D ``[x, y]``. The
state-transition matrix injects ``dt`` into the position-velocity
cross-terms, and ``H`` reads the leading two entries (position).
Process noise is parameterised as ``q * I_4`` and integrated as
per-unit-time noise (see :class:`KalmanLinear`); measurement noise is
``r * I_2``.
Parameters
| Name | Type | Description |
|---|---|---|
| field = 'centroid' | str | Field name for the centroid mean. Covariance lives in ``f"{field}_cov"``. |
| q = 0.01 | float | Per-unit-time process-noise scale. |
| r = 0.1 | float | Measurement-noise scale. |
Members
method
- __call__Advance centroid mean and covariance by one predict step.
- __init__
- make_updateConstruct a matching KalmanUpdate for the 2D centroid observation.
- state_entriesReturn ``(mean, cov)`` :class:`~unitrack.states.State` entries for ``Tracker``.
Source: unitrack/states/kalman/centroid.py:52