Interactive input of measurement values for a Kalman filtering.
The operator sensor_kalman supports the interactive input of measurement values for a Kalman filtering. Kalman filters return an estimate of the current state (or even the prediction of a future state) of a discrete, stochastically disturbed, linear system. They are successfully used in image processing, especially in the analysis of image sequences.
Each filtering is hereby based on certain measurement values. How these values are extracted from images or sensor data depends strongly on the individual application and therefore must be entirely up to the user. However, the operator sensor_kalman allows an interactive input of (fictitious) measurement values y and the corresponding measurement-error covariance matrix R. Especially the testing of Kalman filters during the development can hereby be facilitated.
The parameters MeasurementIn and MeasurementOut include the matrix R which has been stored in row-major order and the measurement vector y lined up, i.e. they are vectors of the length Dimension*Dimension + Dimension
Dimension (input_control) |
integer -> integer |
Number of measurement values. | |
Default value: 1 | |
Range of values: 0 <= Dimension <= 30 |
MeasurementIn (input_control) |
real-array -> real |
The matrix R stored in row-major order and the measurement vector y lined up. | |
Default value: '[1.2,1.0]' | |
Range of values: 0.0 <= MeasurementIn <= 10000.0 |
MeasurementOut (output_control) |
real-array -> real |
The matrix R stored in row-major order and the measurement vector y lined up. |
If the parameters are correct, the operator sensor_kalman returns the value TRUE. Otherwise an exception is raised.
filter_kalman, read_kalman, update_kalman