This operator uses a kalman filter to estimate the distribution of one or more attribute values.
Parameter
- Attributes: The attributes to perform the filter on
- Transition: The transition matrix 'A'
- ProcessNoise: The process noise matrix 'Q'
- Measurement: The measurement matrix 'H'
- MeasurementNoise: The measurement noise matrix 'R'
- InitialState: The initial state vector 'x' (optional)
- InitialError: The initial error matrix 'P' (optional)
- Control: The control matrix 'B' (optional)
Example
out = KALMAN({
ATTRIBUTES = ['x'],
TRANSITION = '[1.0]',
PROCESSNOISE = '[2.0]',
MEASUREMENT = '[1.0]',
MEASUREMENTNOISE = '[4.0]'},
in)
in = KALMAN({
ATTRIBUTES = ['x1','x2'],
INITIALSTATE = '[0.0, 0.0, 0.0, 0.0]',
INITIALERROR = '[1.0,0.0,0.0,0.0;0.0,1.0,0.0,0.0;0.0,0.0,1.0,0.0;0.0,0.0,0.0,1.0]',
TRANSITION = '[1.0,0.0,1.0,0.0;0.0,1.0,0.0,1.0;0.0,0.0,1.0,0.0;0.0,0.0,0.0,1.0]',
PROCESSNOISE = '[1/4, 1/4, 1/2, 1/2;1/4, 1/4, 1/2, 1/2; 1/2, 1/2, 1, 1; 1/2, 1/2, 1, 1]',
MEASUREMENT = '[0.0,0.0,1.0,0.0;0.0,0.0,0.0,1.0]',
MEASUREMENTNOISE = '[10.0,0.0;0.0,10.0]'},
out)