0
$\begingroup$

I have collected force platform data to which I calculate acceleration and velocity. The velocity data is presenting what I understand as drift from the numerical integration process that I undertook to calculate acceleration. Because of this, key time points in the velocity trace are all over the place.

I was therefore wondering if anyone had any recommendations or ideas for which I could minimise/correct the drift I am seeing in my acceleration and subsequent velocity data?

Many thanks in advance, TEli

Improve this question
$\endgroup$
6
  • 1
    $\begingroup$ Telling useful signal from a drift or a trend is a recurrent need. It is important to have an idea on the dynamics of each. Do you have the possibility to dispaly/share some data? $\endgroup$
    – Laurent Duval
    Feb 24, 2022 at 9:16
  • $\begingroup$ Thank you for your reply. Of course, what is the best way of sharing data on here? I am new to stack exchange. I can share later today after work $\endgroup$
    – teli95
    Feb 24, 2022 at 9:54
  • 1
    $\begingroup$ @teli95 that depends on how many data points there are, and what kind of data it is. Generally, things like numpy array (.npy or .npyz) or HDF5 (like your average matlab export) do work fine for many people. If you have a website where you can put arbitrary files (e.g. from your university, employer), that works beatifully. Alternatively, dropbox or wetransfer might do the trick. If this is proper research data that you have the rights to, uploading it to zenodo.org is a very long-term stable way of sharing it. $\endgroup$
    – Marcus Müller
    Feb 24, 2022 at 13:43
  • $\begingroup$ Please see my data attached via the link below. There are a couple of comments to provide clarity. we.tl/t-7veotOIfak $\endgroup$
    – teli95
    Feb 24, 2022 at 21:42
  • $\begingroup$ StackExchange etiquette calls for editing your question until it is complete, and making it stand-alone, with links only for supplementary material such as books or videos that your are citing. If you can make a representative plot that shows your problem, that would be much better than asking people to go to some unknown 3rd-party download site. When you edit your question to include the plot, putting in a link to the download site would be fair, and appreciated by anyone who wants to dig that deep. $\endgroup$
    – TimWescott
    Feb 25, 2022 at 15:18

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Browse other questions tagged or ask your own question.