filter_kalman - Unidimensional Kalman filter, also known as linear quadratic estimation (LQE)
loadrt filter_kalman [count=N|names=name1[,name2...]]
Useful for reducing input signal noise (e.g. from the voltage or temperature sensor).
More information can be found at https://en.wikipedia.org/wiki/Kalman_filter.
Adjusting Qr and Qk covariances:
Default values of Rk and Qk are given for informational purpose only. The nature of the filter requires the parameters to be individually computed.
One of the possible and quite practical method (probably far from being the best) of estimating the Rk covariance is to collect the raw data from the sensor by either asserting the debug pin or using halscope and then compute the covariance using cov() function from Octave package. Ready to use script can be found at https://github.com/dwrobel/TrivialKalmanFilter/blob/master/examples/DS18B20Test/covariance.m.
Adjusting Qk covariance mostly depends on the required response time of the filter. There is a relationship between Qk and response time of the filter that the lower the Qk covariance is the slower the response of the filter is.
Common practice is also to conservatively set Rk and Qk slightly larger then computed ones to get robustness.
filter-kalman.N (requires a floating-point thread)
Update xk-out based on zk input.
filter-kalman.N.debug bit in (default: FALSE)
When asserted, prints out measured and estimated values.
filter-kalman.N.passthrough bit in (default: FALSE)
When asserted, copies measured value into estimated value.
filter-kalman.N.reset bit in (default: FALSE)
When asserted, resets filter to its initial state and returns 0 as an estimated value (reset pin has higher priority than passthrough pin).
filter-kalman.N.zk float in
filter-kalman.N.xk-out float out
filter-kalman.N.Rk float rw (default: 1.17549e-38)
Estimation of the noise covariances (process).
filter-kalman.N.Qk float rw (default: 1.17549e-38)
Estimation of the noise covariances (observation).
Dmian Wrobel <firstname.lastname@example.org>