No announcement yet.

Measuring arm motion

This topic is closed.
  • Filter
  • Time
  • Show
Clear All
new posts

  • Measuring arm motion

    > Date: Sat, 15 Oct 1994 00:02:24 +0100
    > Subject: BIOMCH-L Digest - 13 Oct 1994 to 14 Oct 1994
    > ----------------------------------------------------------------------
    > Date: Fri, 14 Oct 1994 15:15:43 -0300
    > From: WATSON W
    > Subject: Limb Position Measurement
    > Hello;
    > I am a PhD student at the Institute of Biomedical Engineering at
    > the University of New Brunswick in Canada. We work mainly with upper
    > limb prosthetics and my thesis in particular is concerned with
    > end-point-control of a six degree of freedom above the elbow
    > prosthetic arm.
    > The problem that I would like addressed by those who can is the
    > measurement and feedback of humeral, elbow, wrist and hand locations.
    > The wrist will have rotation as well as flexion and extension and the
    > humerus will have rotation. What measurement devices are available on
    > the market? What particular advantages did you discover in using this
    > particular device? What are the advantages and disadvantages with
    > respect to monetary cost, weight, durability and size? Answers to all
    > or some of these questions are greatly appreciated. I will be quite
    > happy to post the answers others may find useful. Thanks.
    > Warren Watson B.A.Sc, M.A.Sc, P. ENG.
    > A1SG@UNB.CA

    Hello Warren,

    To measure motion with as much freedom as you would like, I would
    think the choices are an optical digitizing system with at least
    two cameras, like Optotrak or PEAK or something like the "flock of
    birds" described on BIOMCH-L not too long ago. The latter is a
    6 DofF position sensor based on magnetic flux values, I believe.
    The costs of the three systmes run on the order of $65,000, $45,000
    and $12,000 respectively. Optotrak allows real-time analysis and
    feedback of the signal but no actual video copy. PEAK provides
    video copy but requires digitization separate from the collection
    procedure and therefore can't be done in real time. The "flock of
    birds", I don't know much about, but it sounds intriguing. I believe
    the weight of the sensors is fairly similar across all three devices.
    The Optotrak sensors are active and therefore have wires. PEAK
    uses passive reflective markers and which are entirely unencumbered.

    >From my viewpoint, the most difficult issues are inferring the true
    centers of joint rotation from markers placed on the skin surface,
    and calibrating such data across sessions. As a bioengineer, I'm
    sure you're way ahead of me on the best methods for doing that
    efficiently. My approach is to collect data throughout a reasonable
    range of motion (at least 90 degrees) from three markers per segment
    that are fixed in their spatial relations to each other, and
    calculate the centers of rotation using Woltring's method or some
    such. One does need to collect such data each testing session, as
    it's impossible to place the markers in exactly the same place from
    day to day.

    Feel free to write back if you'd like to discuss any of this further.
    And, I'd like to see what others on BIOMCH-L suggest.

    Rebecca States, Ph.D.
    Dept. of Health & Kinesiology
    Texas A & M University