IEEE Robotics & Automation Magazine - June 2012 - 10

INDUSTRIAL ACTIVITIES

Formulation of a
U.S. National Strategy for Robotics
By Henrik I. Christensen

he robotics industry was
launched in the United States
50 years ago. Since then,
industrial robotics has gained
a lot of popularity. Currently, Japanese and European companies dominate industrial robotics. The domain
has significantly expanded to include
medical, service, and strategic applications. In some of these areas, the
United States still maintains leadership. Some applications are already
mature and new
* ones emerge everyday. Over the last
President Obama
ten years, new
announced the
research and development (R&D)
National Robotics
programs have
Initiative on 23 June
emerged. South
Korea created a
2011 at an event held
national robotics
at Carnegie Mellon
program with significant support.
University.
* The European
Commission started
a major program on cognitive systems. In the United States, there has
been a diverse set of robotics programs funded by different agencies
such as the Defense Advanced Research Projects Agency, National Institute of Standards and Technology, and
the Office of Naval Research. By 2006,
it was apparent that there was no coordinated robotics strategy in the United
States. To address this issue, a group of
senior researchers from most of the
major institutions met for a lunch
Digital Object Identifier 10.1109/MRA.2012.2193931
Date of publication: 6 June 2012

IEEE ROBOTICS & AUTOMATION MAGAZINE

JUNE 2012

meeting at the Robotics Science and
Systems (RSS) Conference in Atlanta
2007. A decision was made to formulate a road map for robotics in the
United States to establish the basis for
a coordinated strategy. For this, a
process was put in place.
Four different workshops were
arranged: 1) manufacturing and logistics, 2) health-care and medical
robotics, 3) service robotics, and 4)
emerging technologies. The first three
workshops had a topical focus, and
the fourth was to consider technology-push issues. For each workshop,
there was a broad call for proposals
(two to three pages). The proposals
were reviewed, and about 30 people
were invited to the workshops. The
meetings had a mixture of representatives from industry, government laboratories, and university researchers.
None of the meetings had the attendance of program managers from funding agencies to avoid focus from any
one agency. The process for each
workshop was to identify
1) major drivers to growth
2) identification of obstacles to growth
3) formulation of R&D to address the
obstacles.
The explicit formulation of the relationship between R&D and growth
was essential to convince others that
new resources for research can have a
direct business impact.
In parallel to the formulation of
topical road maps, a political process
took place. The core team from the
RSS lunch represented 12 universities.
The university presidents formed a
National Leadership Panel chaired by

President Jared Cohen, Carnegie Mellon University, and President Wayne
Clogh (and later, Bud Peterson),
Georgie Institute of Technology. The
leadership in turn was instrumental in
the creation of a congressional caucus
for robotics chaired by U.S. Congressmen Phil Gingrey and Mike Doyle.
The congressional caucus achieved
initial participation from 28 members
of Congress with a balanced participation from both sides of the aisle. The
caucus has later grown to include 35
members of congress. In total, more
than 160 people across industry, government laboratories, and academia
participated in the formulation of the
road map. The main theme coming
out from the different workshops was
the concept of robots working in close
cooperation with humans, which is
referred to as the Co-X concept for
coworkers, coinhabitants, and codefenders. To achieve this vision, there
are a number of obstacles in terms of
lightweight-safe manipulation, robust
navigation, efficient human robot interfaces for nonexperts, and associated technologies, such as robust
perception, formal design, hybrid
control, grasping, HRI, and materials.
Across the three areas of manufacturing, health care and medicine, and
service applications, ten common areas
of research were identified, as shown
in Table 1.
For each area, a general research
agenda was outlined together with
expected targets for five, ten, and
15 years into the future. Major progress in each of these areas would
remove obstacles to further the use of

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - June 2012