Performance
Metrics for Intelligent Systems Workshop - Plenary and Featured
Presentations
PLENARY
ADDRESSES
Prof.
Henrik Christensen
Royal Institute of Technology, Sweden
Georgia
Institute of Technology, USA
Evaluation
of Robots for Human-Robot Interaction
Abstract
Robotics is gradually maturing as a discipline which also implies
an increased need for comparative R&D. At the same time robots
are more and more deployed to serve as assistants to humans be it
for search and rescue or as part of normal daily chores in the home.
To enable evaluation of progress in research it is essential that
rigorous methodologies for evaluation and performance characterisation
are adopted. Often a number of objections are put forward as to
why such rigorous experimental protocols are not well suited for
robotics. Some of the typical objections will be presented and discussed
in the presentation. To illustrate the value and strategy of experimental
evaluation two example applications will be presented. Both applications
are closely tied to robots that serve as assistants to people as
part of daily operations.
A Wizard-of-Oz
study has been used for the design of a strategy for joint human-robot
mapping of domestic setting as part of creation of a cognitive robots
for assistance to people in their homes. The Wizard-of-Oz study
clearly illustrate some of the challenges that a robot has to cope
with and indicate a number of important issues to be considered
in the design of dialog behaviours and associated autonomous functionality.
We will here discuss the use of initial exploratory system designs
to ensure early integration of the end-user.
Another study
has considered the deployment of iRobot PackBot systems for assistance
to soldiers as part of urban intervention. Integration of robots
into a unit of soldiers poses a number of interesting challenges.
There is a here a need to consider the impact of the system on the
robot operator and to re-consider the entire strategy for urban
intervention when a unit is equipped with a light-weight robot for
scouting. Results from a study with the international brigade from
the Swedish military are presented and a number of important lessons
from a long-term (12 month) study is reported.
Experience
from prior studies clearly illustrate the value of a careful design
for evaluation and characterisation of systems, which goes beyond
the simple verification of theoretical models. Observations and
lessons from an extensive set of studies are summarized.
Biography
Henrik I Christensen
is the Kuka Chair of Robotics and a Professor of Computing with
the College of Computing, Georgia Institute of Technology. The appointment
is part-time during 2006, which is a transition period from the
earlier appointment at the Swedish Royal Institutute of Technology
, which included leadership of the Center for Autonomous Systems.
He does research on mobile robotics, autonomous systems, computer
vision, and biologically inspired robot systems. The overall emphasis
is on a holistic approach to design of systems, incl mathematically
well defined methods for design, analysis and implementation of
systems. A fundamental idea is that methods should be evaluated
in realistic settings which involves an interesting scenario and
a full systems context. He is involved in a large number of national
and international projects. Dr. Christensen is a co-founder of the
company Intelligent Machines and serve as a scientific advisor to
Evolution Robotics. Research cooperation involves research labs
and companies on three continents. In addition he has been actively
involved in a number of community efforts in particular as the founding
coordinator of the EU network of excellence in Robotics - EURON
(2000-2006). Dr. Christensen is a fellow of the International Foundation
of Robotics Research and served as an IEEE RAS distinguished lecturer
(2004-2006). He
also serves on the board of trustees of the Swedish STINT foundation.
[Return
to top]
Prof.
Shigeo Hirose
Tokyo Institute of Technology, Japan
Development
of Rescue and Demining Robots in Tokyo Institute of Technology
Abstract
In this plenary talk, I will explain about our activities on rescue
and demining robots. As for the robots for rescue operation, I will
first explain my previous efforts on snake-like robots with slender
and actively bending bodies. I will then show several types of snake-like
"Soryu" robots which consist of three crawler-driven segments
and their connecting joints. The Soryu has been adapted with a specific
driving mechanisms to move inside narrow and winding paths among
debris and is designed to protect against dust and water. A newly
introduced crawler belt made of thin metal with rubber knobs will
also be explained. I will also present a debris-inserting inspection
camera, we are developing with a snake-like expandable rod mechanism.
In general, I will introduce our development process for these and
other devices. We believe that the most effective rescue tools will
be the ones which are widely used in our daily life. Based on this
belief, we also paid special attention to the development of ordinary-life-embedded
rescue devices. For example, automobile jack-up devices which can
be used for rescue operations will be shown. As for the demining
robots, I will explain about my preliminary efforts to develop walking-demining
robots, and their tool-detachable foot mechanisms. I will explain
about our latest activities on a practical demining vehicle named
"Gryphon." It has a weight balanced arm with metal and
ground penetrating radar and a 3D camera. It can measure the uneven
ground and can drive the sensors along the surface of the ground.
I will show the result of the experiments in several places such
as in Croatia.
Biography
Shigeo Hirose was born in Tokyo in 1947. He received the B. E. degree
with first class honors in Mechanical Engineering from Yokohama
National University in 1971, and his M. E. and Dr. E. degrees in
Control Engineering from the Tokyo Institute of Technology in 1973
and 1976, respectively. He was Research Associate and Associate
Professor of the same university, and since 1992 he has been a Professor
of Tokyo Institute of Technology, Department of Mechanical and Aerospace
Engineering. He is a Fellow of IEEE, JSME and RSJ. His research
interest is in the creative design of robotic mechanisms and their
control. He has been awarded more than 30 academic prizes including
the "Medal with Purple Ribbon" from the Japanese government
(2006), the first Pioneer in Robotics and Automation Award (1999),
and the Best Conference Paper Award (1995) from the IEEE Robotics
& Automation Society.
[Return
to top]
Prof.
Hugh Durrant-Whyte
ARC
Federation Fellow, Research Director
ARC Centre of Excellence for Autonomous Systems
The University of Sydney, Australia
hugh@cas.edu.au
Maximal
Information Systems
Abstract
Information provides a quantitative metric for describing the value
of individual systems components in autonomous systems tasks such
as tracking, mapping and navigation, search and exploration; tasks
in which the objective is information gain in some form. An information
model is an abstraction of system capabilities in an anonymous form
which allows a priori reasoning on the system itself. By construction,
information measures have properties of composability and additivity
and thus provides a natural means of modelling and describing large
scale systems of systems.
This talk will
begin by describing how information measures arise naturally in
autonomous tracking, mapping and navigation, search and exploration
tasks. It is then demonstrated that the performance of individual
sensors and platforms can be modelled using these information measures
and that system-level performance metrics can be computed. These
ideas are illustrated in a series of tasks involving mixed air and
ground autonomous systems. These include flight-tests of cooperative
UAVs engaged in tracking and navigation tasks, mixed UAV, ground
vehicles and human operatives, engaged in mapping and picture compilation
operations, and operations involving multi-platform search in constrained
environments. In each, it is shown how information provides both
a performance metric and design objective underpinning large-scale
systems of systems operation.
Biography
Hugh Durrant-Whyte received the B.Sc. in Nuclear Engineering from
the University of London, U.K., in 1983, and the M.S.E. and Ph.D.
degrees, both in Systems Engineering, from the University of Pennsylvania,
U.S.A., in 1985 and 1986, respectively. From 1987 to 1995, he was
a Senior Lecturer in Engineering Science, the University of Oxford,
U.K. and a Fellow of Oriel College Oxford. From 1995 to 2002 he
was Professor of Mechatronic Engineering at University of Sydney.
In 2002 he was awarded an inaugural Australian Research Council
(ARC) Federation Fellowship. He also now leads the ARC Centre of
Excellence in Autonomous Systems. His research work focuses on autonomous
vehicle navigation and decentralised data fusion methods. His work
in applications includes automation in cargo handling, mining, defence,
and marine systems. He has published over 300 technical papers and
has won numerous awards and prizes for his work. He is a Fellow
of the Academy of Technical Sciences, a Fellow of the IEEE and an
IEEE Robotics Society Distinguished Lecturer.
[Return
to top]
Dr.
Martin Buehler
Boston
Dynamics, USA
Developing
Dynamic Legged Robots - Towards Greater Mobility Without Falling
Over
Abstract
Mobility can
be an important contributor to robot intelligence, for gathering
information, implementing decisions, and interacting with the environment.
While wheeled and tracked robots have a relatively easy time moving
around, we have to invest some intelligence first into legged robot
design and control in order to harvest their potentially much greater
mobility.
This talk will
describe several recent legged robots that walk, run, balance, climb,
carry loads, resist kicks and negotiate rough terrain with new levels
of dynamic mobility, robustness, and performance. In the process
we will encounter interesting issues related to the system design,
performance metrics, energy efficiency, and the experimental evaluation
of these systems.
Biography
Martin Buehler
received the M.Eng. and Ph.D. degrees in Electrical Engineering
from Yale University in 1985 and 1990. His doctoral work focused
on the design, control and analysis of juggling robots and the analysis
of a hopping robot. After a Postdoc at MIT's leglab on dynamic legged
locomotion, he joined McGill University, Montreal, in 1991 as an
NSERC Junior Industrial Research Chair and a Scholar of the Canadian
Institute for Advanced Research. He founded and headed the Ambulatory
Robotics Lab, which produced one, four and six legged robots, including
the ARL Monopods I and II, Scout I and II, CARL, PAW, RHex and AQUA,
funded by major Canadian government, DARPA and industrial contracts
and grants. In 2003 he received McGill's William Dawson Scholar
Award. In the same year he moved on to become Director of Robotics
at Boston Dynamics, Cambridge, USA. Dr. Buehler served as an Associate
Editor of the IEEE Transactions on Robotics and Automation from
1998 - 2003, and is currently on the editorial boards of the International
Journal of Robotics Research and the Journal of Field Robotics.
He has supervised over 30 graduate students at McGill and has published
over 100 papers on legged robot design and control, dynamic manipulation
and motor control.
[Return
to top]
Dr.
James Albus
National Institute of Standards and Technology, USA
Building
Brains for Thinking Machines
Abstract
In this talk,
Dr. Albus will describe how research in computer science, control
theory, and the neurosciences are converging towards intelligent
systems that can mimic human performance in a broad range of applications.
He will discuss current efforts to build machines that can perceive
the environment, build an internal model of the external world,
and use that model for decision-making, reasoning, planning, and
real-time control of complex machines in uncertain, and potentially
hostile, environments. He will suggest how system architectures
designed for autonomous mobility systems are computationally similar
in many respects to the human brain, and vice versa.
This work is
part of a broad NIST program of research and engineering of intelligent
systems to reduce costs and improve quality in manufacturing and
construction, and to save lives of civilians on the highway and
soldiers in combat. The research is conducted in collaboration with
the Army Research Laboratory, DARPA, the Department of Transportation,
and the U.S. manufacturing industry.
Bio
Dr. James S. Albus founded and led the Intelligent Systems Division
at the National Institute of Standards and Technology for 20 years.
He is currently a Senior NIST Fellow. Over a long and varied career
Dr. Albus has made a number of scientific contributions. During
the 1960's he designed electro-optical systems for more than 15
NASA spacecraft. During the 1970's, he developed a model of the
cerebellum that after 30 years is still a leading theoretical model
used by cerebellar neurophysiologists today. Based on that model,
he invented the CMAC neural net, and co-invented the Real-time Control
System (RCS). RCS is a reference model architecture for intelligent
systems that has been used over the past 25 years for a number of
systems including the NBS Automated Manufacturing Research Facility
(AMRF), the NASA telerobotic servicer, a DARPA Multiple Autonomous
Undersea Vehicle project, a nuclear Submarine Operational Automation
System, a Post Office General Mail facility, a Bureau of Mines automated
mining system, commercial open architecture machine tool controllers,
and numerous advanced robotic projects, including the Army Research
Lab Demo III Experimental Unmanned Ground vehicle. The latest version
of the RCS architecture has been selected by the Army for the Autonomous
Navigation Systems to be used on all Future Combat System ground
vehicles, both manned and unmanned. He is also the inventor of the
NIST RoboCrane. He is currently working with DARPA and other government
agencies on a concept for a National Program for Understanding the
Mind, a.k.a "Decade of the Mind."
Dr. Albus has
received numerous awards for his work in control theory including
the NIST Applied Research Award, the Department of Commerce Gold
and Silver Medals, the Industrial Research IR-100 award, the Presidential
Rank Meritorious Executive, the Jacob Rabinow award, the Japanese
Industrial Robot Association R&D Award, and the Joseph F. Engelberger
Award for robotics technology. In 1998, he was named a "Hero
of Manufacturing" by Fortune magazine.
Dr. Albus is
the author of more than 180 scientific papers, journal articles,
book chapters, and official government studies on intelligent systems
and robotics. He has lectured extensively throughout the world and
authored or co-authored five books:
Engineering
of Mind: An Introduction to the Science of Intelligent Systems
- Wiley, 2001
Intelligent Systems: Architecture, Design, and Control
- Wiley, 2002
The RCS Handbook: Tools for Real-Time Control Systems Software
Development - Wiley, 2001
Brains, Behavior, and Robotics - Byte/McGraw-Hill,
1981
Peoples' Capitalism: The Economics of the Robot Revolution
- New World Books, 1976
He is a member
of the editorial board of the Wiley Series on Intelligent Systems
serves on the editorial boards of six journals related to intelligent
systems and robotics.
Dr. Albus received
a B.S. in Physics from Wheaton College (Illinois) in 1957, a M.S.
in Electrical Engineering from Ohio State University in 1958, and
a Ph.D. in Electrical Engineering from University of Maryland (College
Park) in 1972.
[Return
to top]
FEATURED
PRESENTATIONS
Mr.
Chuck Shoemaker
Robotic Research LLC, USA
Army
Science and Technology Initiatives for Autonomous Tactical UGVs:
The Last 10 Years
Abstract
The U.S. Army has focused growing attention and resources on the
development of the Science and Technology Base required to make
Autonomous UGVs a tactical reality. During the last 10 years a few
of the Basic and Applied Research programs in this area have demonstrated
capabilities and maturity levels leading the Army to invest major
resources to accelerate the technology maturation required for initial
deployment of this technology. This paper addresses a number of
the critical efforts that have been the linchpins of this technology
evolution.
[Return
to top]
Dr.
Mike Montemerlo
Stanford University, USA
Winning
the DARPA Grand Challenge
Abstract
The DARPA Grand Challenge has been the most significant event for
the robotics community in more than a decade. A mobile ground robot
had to traverse 132 miles of unrehearsed desert terrain in less
than 10 hours. In 2004, the best robot only made 7.3 miles. In 2005,
Stanford won the challenge and the $2M prize in less than 7 hours
travel time, and ahead of four other finishers. This talk, delivered
by the Software Lead developer of the Stanford Racing Team, will
provide insights into the software architecture of Stanford's winning
robot. The robot massively relied on machine learning and probabilistic
modeling for sensor interpretation and control. The speaker will
explain some of the basic algorithms that made this victory possible,
and share some of the excitement characterizing this historic event.
[Return
to top]
INVITED
TALKS
