You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 5 Next »

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)


  • No labels