PDA

View Full Version : acceleration errors



Christian Lariviere
03-07-1996, 05:39 AM
> Hello Biomch-l readers:
>
> I am posting the responses I received to the following question:
> I don't know what is the magnitude of errors concerning
>accelerations of segments in human movement tasks when
>displacement data are differentiated twice to get accelerations. Is
>there any references or ideas that could help me to estimate this
>error.
> I would sincerely like to thank all of those individuals for
>responding to this question.
> Sincerely, Christian Lariviere

************************************************** *******
> Subject: Re: acceleration errors
> To: CLARIVIERE@FEPS.USHERB.CA
> Date: Thu, 22 Feb 96 18:36:45 MST
> From: "Ton van den Bogert"

> Christian,
>
> First, displacement data must me low-pass filtered to avoid
> excessive noise in the accelerations.
>
> To estimate the error in acceleration, after filtering, you can
> use Lanshammar's formula. This assumes a band-limited signal
> with added white noise and a low-pass filter which passes all
> signals within the bandwidth of the signal.
>
> Reference:
>
> H. Lanshammar (1982) On precision limits for derivatives
> numerically calculated from noisy data. J. Biomechanics
> 15:459-470.
>
> You might get a reply from Lanshammar, since he subscribes to
> this list.
>
> -- Ton van den Bogert
> Human Performance Laboratory
> University of Calgary
>

************************************************** ************

> Date: Thu, 22 Feb 1996 22:48:54 -0800
> From: ariel1@ix.netcom.com (Gideon Ariel )
> Subject: Re: acceleration errors
> To: Christian Lariviere

> Surf the net at URL: http://www.arielnet.com

************************************************** ************
> Date: Sun, 25 Feb 96 15:40:47 -0400
> From: ED BIDEN
> To:
> Subject: Re: acceleration errors

> Hi:
>
> A good place to start is to read "An introductionto Error Analysis"
> by John Taylor (about 1981). I have a student who has done some
> similar calculations. You should request Chris. McGibbon's thesis
> "Sensitivity of Analytical Knee Models to Input Data Errors" from
> the University of New Brunswick library system, or CISTI. Chris
> has been busy at other things and it is not yet published.
>
> Cheers,
>
> Ed Biden
> Inst. of Biomed Engg
> Univ. of NB
************************************************** ************