IEEE Robotics & Automation Magazine - June 2012 - 73

Jet Propulsion Lab (JPL), the University of Illinois, and
some institutions in Japan and Europe. The publication
venues were few. Information was exchanged through visits and technical reports more often than through journals
and conferences. The conferences that focused on robotics
were RoManSy, held biannually in Europe, and the International Symposium on Industrial Robots, and the only
journal that focused on robotics was Industrial Robot.
My training was informal, guided primarily by three
senior graduate students: Marc Raibert, John Hollerbach,
and Tomas Lozano-Perez (later my coadvisor). A few of
us formed a study group and conceived the idea of publishing an edited collection of papers to address the poor
state of robotics literature. We were joined by Mike
Brady and Tim Johnson and wrote Robot Motion: Planning and Control, which was ultimately published in
1983 and commonly referred to as the Brady book. It
included some tutorial material as well as papers in kinematics and dynamics, control, force control, trajectory
planning, and spatial planning.
The significance of the Brady book to this essay was its
inclusion of planning along with the more familiar topics
of kinematics, dynamics, and control. This marked a
change in the role of computer science in robotics. The
1970s had seen the development of programming languages for robotic manipulators and interest in high-level
languages and automatic programming. A notable example was the development of SRI's autonomous mobile
manipulator Shakey [5], [20], applying emerging automated theorem proving techniques in robot planning.

Even though these accomplishments were impressive,
robot-motion planning did not flourish until a geometric
approach was developed in the 1980s. The perspective of
the 1970s is preserved in the seminal textbook Robot
Manipulators: Mathematics, Programming, and Control
written by Paul [21]. There is no chapter on planning.
Instead, the final chapter is "Programming."
In the late 1970s, the focus on programming
languages was displaced by a geometric perspective,
contemporary with the
appearance of a new field: *
computational geometry.
Information was exchanged
Lozano-Perez and Wesley developed the use of
through visits and technical
configuration space for
collision-free
motion
reports more often than
planning in 1977 [16],
[17]. I applied the apthrough journals and
proach to compliant control in 1978 [18], [19].
conferences.
The problems of geo- *
metric motion planning
crossed over to computational geometry, with papers by
Reif in 1978 [23] and Schwartz and Sharir [26]. When Brady
and Paul launched the first robotics journal in 1982 [the
International Journal of Robotics Research (IJRR)], the solicitation of papers explicitly included the topics spatial planning and world modeling. Many of us adopted a new
definition of robotics: the intelligent connection of perception to action [1]. Robotics has emerged as a research field.

Goertz
etc.

Telerobotics

Unimate
etc.

Industrial Robotics

RoManSy
etc.

Mechanical Science

IJRR
etc.

Computer Science

ior

v
ha

Creations
1940

1950

1960

1970

ica

ist

t
ta

Revolutions
1980

1990

2000

2010

Figure 1. The creations of robotics.

JUNE 2012

IEEE ROBOTICS & AUTOMATION MAGAZINE

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