Page History

This operator uses a kalman filter to estimate the distribution of one or more attribute values Kalman Filter to produce a statistically optimal estimate of the underlying system state.

Parameter

• Variables: The name of the variables
• Attributesattributes: The attributes to perform feed the filter on
• transitionTransition: The transition matrix 'A'
• controlProcessNoise: The control process noise matrix 'Q'processnoise
• Measurement: The process noise measurement matrix 'H'
• measurementMeasurementNoise: The measurement noise matrix
• 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({
              VARIABLES = ['x'], 
              ATTRIBUTES = ['m'],
              TRANSITION = '[1.0]', 
              PROCESSNOISE = '[2.0]',
              MEASUREMENT = '[1.0]', 
              MEASUREMENTNOISE = '[4.0]'}, 
             in)

outputout = kalmanfilterKALMAN({attributes
             VARIABLES = ['x','y','dx','dy'], 
             ATTRIBUTES = ['vx','vy'], 
             INITIALSTATE = '[0.0, 0.0, 0.0, 0.0]', transition=[], control=[], processnoise=[], measurement=[], measurementnoies=[]}, input
             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]'},
            in)