Tweet
Dear BIOMCH-L readers:
A team of us at the University of British Columbia are
developing a powered upper limb orthosis (it could be compared
to a robotic arm into which a person's paralyzed are would be
placed) so that people can regain the ability to eat, reach for
objects etc.
First we interviewed potential users to figure out what
they would most want to do (if you try to do everything the
device gets too complex, costly, prone to breakdown etc.) in
order to attach a priority to various tasks. Next we'll be
doing a motion analysis study of the top-priority tasks. Then
I'll be feeding that information into a kinematic simulation
program to test out various configurations of an orthosis. The
motion analysis uses videocameras, reflective markers, and software
adopted from a University of Manitoba program (R. Safee-Rad, E. Schwedyk,
A. Quanbury). The simulation program is designed such that we can
keep reducing the degrees-of-freedom while still being able to perform
the selection of tasks. DOF can be reduced by fixing it (e.g. by
keeping the arm at a fixed distance out from the body) or by combining
it with others (e.g. rotating the forearm as the elbow bends). The
ultimate goal, then, is to analytically determine the "optimal"
configuration for a powered upper limb orthosis. We'll then apply
this to a new design.
My questions are: 1) Has anybody else been studying the motion of
the entire arm, aside from Safee-Rad et al., Engen, Langrana or McWilliam?
2) What problems have people encountered with 2 cameras vs. 3?
3) Has anybody generated the workspace volume of the human arm -
preferrably with options to limit various joints, in comparison to a
required workspace volume, or to be able to reach particular points: I've
developed a complete program, but it's not necessarily that "elegant"!
Thanks in advance for your help,
Carolyn Anglin
Dept. of Mechanical Engineering
University of British Columbia
2324 Main Mall
Vancouver, B.C., Canada
V6T 1Z4
email: Carolyn_Anglin@mtsg.ubc.ca
Tel: (604) 822-7862
FAX: (604) 822-2403
A team of us at the University of British Columbia are
developing a powered upper limb orthosis (it could be compared
to a robotic arm into which a person's paralyzed are would be
placed) so that people can regain the ability to eat, reach for
objects etc.
First we interviewed potential users to figure out what
they would most want to do (if you try to do everything the
device gets too complex, costly, prone to breakdown etc.) in
order to attach a priority to various tasks. Next we'll be
doing a motion analysis study of the top-priority tasks. Then
I'll be feeding that information into a kinematic simulation
program to test out various configurations of an orthosis. The
motion analysis uses videocameras, reflective markers, and software
adopted from a University of Manitoba program (R. Safee-Rad, E. Schwedyk,
A. Quanbury). The simulation program is designed such that we can
keep reducing the degrees-of-freedom while still being able to perform
the selection of tasks. DOF can be reduced by fixing it (e.g. by
keeping the arm at a fixed distance out from the body) or by combining
it with others (e.g. rotating the forearm as the elbow bends). The
ultimate goal, then, is to analytically determine the "optimal"
configuration for a powered upper limb orthosis. We'll then apply
this to a new design.
My questions are: 1) Has anybody else been studying the motion of
the entire arm, aside from Safee-Rad et al., Engen, Langrana or McWilliam?
2) What problems have people encountered with 2 cameras vs. 3?
3) Has anybody generated the workspace volume of the human arm -
preferrably with options to limit various joints, in comparison to a
required workspace volume, or to be able to reach particular points: I've
developed a complete program, but it's not necessarily that "elegant"!
Thanks in advance for your help,
Carolyn Anglin
Dept. of Mechanical Engineering
University of British Columbia
2324 Main Mall
Vancouver, B.C., Canada
V6T 1Z4
email: Carolyn_Anglin@mtsg.ubc.ca
Tel: (604) 822-7862
FAX: (604) 822-2403