3
$\begingroup$

Here is a set of altitude data I collected with an Arduino sensor BMP-180 on a board CanSat

data = {13.15, 13.4, 13.15, 13.31, 13.4, 13.31, 13.15,.....}

how can I use Kalman Filter in Mathematica to do the prediction on that set of data?

$\endgroup$
  • 2
    $\begingroup$ Does your question relate to the software Mathematica or to the science of mathematics. If the latter, I suggest that you move your question to Mathematics. $\endgroup$ – bbgodfrey Sep 25 '15 at 15:42
  • 1
    $\begingroup$ ... or maybe even Signal Processing $\endgroup$ – sebhofer Sep 25 '15 at 16:03
  • 3
    $\begingroup$ ... and next to data the good old question remains: "What is your model?" Have a look at the documentation for KalmanFilter. $\endgroup$ – gwr Sep 25 '15 at 16:07
  • $\begingroup$ @gwr ...which is probably much harder to answer than the original question... $\endgroup$ – sebhofer Sep 26 '15 at 11:33
7
$\begingroup$

Something like this perhaps.

tsm = TimeSeriesModelFit[data];
ListLinePlot[{tsm["TemporalData"],
  TimeSeriesForecast[Normal[tsm], data, {6}, Method -> "Kalman"]}]
$\endgroup$
  • $\begingroup$ Thanks! May I also know how I can get the position of a certain value through the prediction? $\endgroup$ – Yincheng Liu Sep 27 '15 at 13:04
  • $\begingroup$ For example: Select[TimeSeriesForecast[Normal[tsm], data, {30}, Method -> "Kalman"]["Path"], Last[#] <= 13 &] $\endgroup$ – Chris Degnen Sep 27 '15 at 14:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.