IEEE Robotics & Automation Magazine - June 2011 - 125

*
could, for example, bring in the threedimensional (3-D) computer-aided
design (CAD) model of a car body and
investigate where to place the robot so
that it could reach all the spot-weld
points, how the spot-weld gun should
be mounted on the flange, work out all
the details, make sure the selected
robot could do the job, avoid collisions,
and so on. Then, the offline programming system would create an online
program and download it to the actual
machine. To make all this actually
work, it got me into the problem of
robot calibration, and of course, also
learned a lot about 3-D modeling,
graphics, and fast collision detection
algorithms. Our timing was good, and
when we started, graphics were pretty
wimpy, and each link of the robot was
depicted as a single line segment. But
within a year or two, the first Silicon
Graphics workstations were out, and
we had beautiful 3-D surfaced models
of the robot links, the tooling, and the
work pieces. We were at the right time
to put together practical systems that
tied CAD models to automated manufacturing. I was convinced at that time
that we really didn't need faster or
more-accurate robots; we needed easierto-use robots so that they could be cost
effectively deployed into factories. So,
that's what we tried to do.

PIC:

How did you come to work
for Adept?
JJC: Adept acquired Silma in 1995,
and we became the Silma Division of
Adept. I got involved in some robot
design issues for the main business of

n "RAS Finances," IEEE Robotics &
Automation Magazine, volume 18,
number 1, March 2011, pp. 8-10,
the text at the end of the first

Digital Object Identifier 10.1109/MRA.2011.941639
Date of publication: 14 June 2011

Adept, but mostly I worked in the Silma
division and created a new version of
the offline-programming and simulation system that was specific to the sort
of applications that Adept robots were
used in. I was happy to work with Brian
Carlisle, Bruce Shimano, Rick Casler,
and others. Silma existed for about ten
years before being acquired by Adept,
and I stayed with Adept for an additional seven years after the acquisition,
so all together, I spent 17 years in the
area of CAD-driven offline programming of robots.

PIC: What came next?
JJC: After Adept, I worked

for a
high-tech company called Invenios.
They make high-precision robots-
generally intended for work on very
small-sized applications-to assemble
fiber optics, as stages for atomic-force
microscopes, and other small things.
One project there was a 6-DoF parallel
robot for which I had to develop an
iterative kinematic algorithm, because
the geometry was not closed-form
solvable. It could position within tens
of micrometer and had a workspace
smaller than a cubic centimeter. I also
developed a digital signal processing
(DSP)-based encoder that was pretty
advanced.

PIC: Can you tell me what you are
up to now?
JJC: For the past two years, I have been
an entrepreneur-my wife and I ran
our two companies: one is a software
company (www.Metron-Imaging.com)
that makes a software package used

for veterinary radiography; the other
company is a horseshoe company
(www.EponaShoe.com), and we make
polyurethane horseshoes that are better for the horse than the 2,000-yearold metal technology still widely used.
Our shoes were on the top dressage
horse on the U.S. Olympic team in Beijing. It may sound like a big change for
me, but basically I've just shifted from
robot mechanics to the related field of
biomechanics, with a special focus on
the horse. We've owned horses for 20
years and have 12 horses now, and so,
over time I guess I got interested in
how they work.

PIC: Finally, what do you see as the
big opportunities and challenges for
robotics today?
JJC: I'm pretty excited about all the
work in the area of haptics and robots
than can control their applied forces
and work safely alongside humans. I
think there are more applications to
come in that area.
I guess I would *
Markets for robots
also still say (as I
did 25 years ago)
would further open up
that making robots
if they were less costly
easier to use is a
key thing. This
to deploy.
means either they
*
become way easier to program
or they become more autonomous.
Markets for robots would further
open up if they were less costly to
deploy-and that comes from them
being easier to use.

ERRATA

paragraph of the "Budget Year
2011 and Summary" section should
have been "In the past several years,
the budget surpluses were À$78.9K,
À$46.2K, À$42.5K, À$59.1K, $78.1K
in 2006, 2007, 2008, 2009, and 2010,
respectively."

These surplus numbers appeared
incorrectly; they inadvertently had
been shifted by one decimal place.
For example, À$46.2k was printed as
ÀUS$462,000. We apologize for the
error and any confusion it may have
caused.

JUNE 2011

IEEE ROBOTICS & AUTOMATION MAGAZINE

125

http://www.EponaShoe.com http://www.Metron-Imaging.com

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - June 2011