|
|
|
|
9:30-10:20, November 1
"Robotics with LEGO MINDSTORMS"
Prof. Henrik Hautop Lund, the Maersk Institute at University of Southern
Denmark, Denmark
9:30-10:20, November 2
"Robotics as a Computational
Neuroscience"
Dr. Mitsuo Kawato, ATR HIP Dept. 3, ATR International CyberHuman Project,
ERATO JST, Japan
9:30-10:20, November 3
"Medical
Robotics and Computer-Integrated Therapy Delivery: Coupling Information
to Action in 21'st Century Surgery"
Prof. Russell H. Taylor, John Hopkins University, USA
Abstract:
When designing new robotic systems, the LEGO MINDSTORMS platform offers
the possibility to make fast prototyping, since robots with radically different
designs can be constructed in very short time. We have developed new tools
based on adaptive robotics to allow both expert and non-expert users to
develop robust and reliable LEGO MINDSTORMS robots within an hour. These
techniques include user-guided behavior-based robotics, user-guided evolutionary
robotics, and co-evolutionary robotics. But, when designing task-fulfilling
robots, we must be careful not to focus exclusively on the robot control.
Instead of looking only at the development of robot controllers (as is
the standard way of looking at this problem) I will take another view and
argue that it is necessary to develop both the body and the brain in order
to achieve the best performing robots (e.g. through simulated evolution).
This is in accordance with new considerations within the field of embodied
artificial intelligence, as for instance the ecological balance principle.
The use of such a Building Brains and Bodies study is facilitated by the
use of the LEGO MINDSTORMS Robotics Invention System that allows easy assembling
and prototyping of robots with different morphologies. Further, I will
show how the behavior-based approach can be used to develop LEGO robot
soccer games. This can be done either with a distributed behavior-based
approach, or with a more straight forward behavior-based approach that
allows children in the age 7-14 to successfully develop their own robot
soccer players to participate in robot soccer tournaments, such as RoboCup
Jr.
Speaker Biography:
Henrik Hautop
Lund is professor in engineering at The Maersk Institute at University
of Southern Denmark. Since 1997, he is head of The LEGO Lab The LEGO Lab
is a young lab financed by the Danish Government and Danish industries,
such as LEGO. Currently, the LEGO Lab is occupied with development of LEGO
MINDSTORMS robots and how to put artificial intelligence into our daily
life.
Henrik Hautop Lund has previously worked as research associate for a couple of years (during the period 1992-1995) at the Institute of Psychology, The National Research Council, Rome, Italy, doing research in the fields of artificial life, neural networks, and evolutionary computation. Further, Henrik Hautop Lund has worked as research associate for a couple of years (1996-1997) in the Department of Artificial Intelligence at University of Edinburgh, UK, doing research on biologically inspired robotics and evolutionary robotics. He has been guest editor of journals such as the MIT Press Artificial Life journal. He has published approximately 40 peer reviewed papers in international, scientific journals and conference proceedings.
His research has been featured as cover stories of magazines such as Business Week and New Scientist, and robots from his lab has been featured in numerous TV shows (including CNN, Azteca, NHK, Reuters, AP, etc.)
Info and publications:
http://www.daimi.au.dk/~hhl/
LEGO Lab:
http://legolab.daimi.au.dk
develop their own robot soccer players to participate in robot soccer
tournaments, such as RoboCup Jr.
Abstract:
I define the science discipline "computational neuroscience" as follows.
We investigate information processing of the brain to the extent that artificial
machines, either computer program or robot can solve the same computational
problems, solved by the brain, essentially in the same principle.
This new discipline of the neuroscience is essential for understanding
mechanisms of information processing in the brain. This is partly
because previous experimental and biological neuroscience studies advanced
enormously for the last 40 years but they accumulated data mainly on the
information about the substances in the brain or functional localization
within the brain. These data are valuable but not sufficient at all
for reproducing or understanding information processing in the brain.
In other words, we need to create a brain if we really want to understand
brain mechanisms in deep senses. However, brain functions can not
be studied dealing with only the brain. We also need to reproduce
bodies and surrounding environments. Then, it is obvious that robotics
research is very much related to the above-defined computational neuroscience.
In the past, this scientific aspect of robotics to elucidate information
processing of human intelligence has not been emphasized. Furthermore,
unfortunately, this scientific objective of robotics was even hidden, made
implicit or neglected in some engineering frameworks. On the contrary,
I strongly believe that the most important aim of at least humanoid robotics
is elucidating human intelligence. Accordingly, for more than 10 years,
our groups at ERATO and ATR have been pursuing this scientific aspect of
robotics as a new branch of the computational neuroscience. We have
been investigating many different topics such as Imitation learning, computational
learning theory, eye movement neural network, neural oscillators. We used
several robots in the past but recently concentrate on our humanoid robot
DB (http://www.erato.atr.co.jp/DB/). DB stands for Dynamic Brain.
DB has 30 degrees of freedom for eye, head, arms, body trunk and legs,
and possesses hydraulic actuators for these freedoms. DB has four
cameras, joint angle sensors and force sensors for all the actuators.
It is compliant and very quick in movements. For a review on our
recent efforts using DB, we refer to (Atkeson CG, Hale J, Kawato M, Kotosaka
S, F. Pollick, Riley M, Schaal S, Shibata T, Tevatia G, Ude A, Vijayakumar
S: Using humanoid robots to study human behavior. IEEE Intelligent Systems:
Systems Special Issue on Humanoids, in press (2000)).
Speaker Biography:
Name:
Mitsuo Kawato
Date of Birth: November 12, 1953.
Education:
1976 B.S. degree in physics from The University of
Tokyo
1978 M.E. degree in biophysical engineering from
Osaka University
1981 Ph. D. degree in biophysical engineering from
Osaka University
Professional Positions:
1981 Faculty member of Osaka University
1987 Lecturer of Biophysical Engineering, Faculty
of Engineering Science, Osaka University.
1986 -1987 Jointly appointed as a senior
science advisor of Ministry of Education, Culture and Science of Japan.
1988 Senior researcher of Cognitive Processes Department,
ATR Auditory and Visual Perception Research Laboratories
1989 Supervisor of Cognitive Processes Department,
ATR Auditory and Visual Perception Research Laboratories
1992 Head of department head of Department 3, ATR
Human Information Processing Research Laboratories.
1996 Jointly appointed as a leader of Kawato Dynamic
Brain Project, ERATO, JST.
1992 -1995 Jointly appointed as a visiting
professor of Laboratory for Parallel Distributed Processing, Institute
for Electronic Sciences of Hokkaido University.
1993 Jointly appointed as a professor a Contratto
di Robitica of University Genova
1994 Jointly appointed as visiting professor of Kanazawa
Institute of Technology.
2000 Jointly appointed as visiting professor of Nara
Institute of Science and Technology
Activities:
Governing board member of Japan Neural Network
Foreign member of North America Society for Neuroscience
Member of Japan Institute of Electronics, Information and Communication
Engineers.
Co-editor-in-chief of Neural Networks.
For the last fifteen years he has been working in computational neuroscience and neural network modeling. He published about 200 papers, reviews and books. Research topics include simulation study of dendritic spines, feedback-error-learning model and its applications to industrial robot manipulators, movement trajectory formation, bi-directional theory for interactions between cortical areas, cerebellar internal models, and teaching by demonstration for robots.
Awards:
1986 Toyama award of The Toyama Foundation
1988 Grant-in-aid award of Brain Science Foundation
1991 Excellent paper award and Yonezawa founder's
medal memorial special award of The Institute of Electronics, Information
and Communication Engineers
1992 Sawaragi memorial paper award of Society of
Instrument and Control Engineers Outstanding research award of the
International Neural Network Society
1993 Persons of scientific and technological research
merits of commendation by the Ministry of state for Science and Technology
1993 Paper award and research award of Japan Neural
Networks Society Osaka Science Prize
1994 Paper award and research award of Japan Neural
Networks Society
1996 10th Tsukahara Naka-akira Memorial Award
1997 Paper award of Japan Neural Networks Society
Okawa Publications Prize
1998 Tomoda paper award of the Society of Instrument
and Control Engineers
Abstract:
The impact of Computer-Integrated Surgery (CIS) on medicine in the
next 20 years will be as great as that of Computer-Integrated Manufacturing
on industrial production over the past 20 years. A novel partnership between
human surgeons and machines, made possible by advances in computing and
engineering technology, will overcome many of the limitations of traditional
surgery. By extending human surgeonsf ability to plan and carry
out surgical interventions more accurately and less invasively, CIS systems
will address a vital national need to greatly reduce costs, improve clinical
outcomes, and improve the efficiency of health care delivery. As
CIS systems evolve, we expect to see the emergence of two dominant and
complementary paradigms: Surgical CAD/CAM systems will integrate accurate
patient-specific models, surgical plan optimization, and a variety of execution
environments permitting the plans to be carried out accurately, safely,
and with minimal invasiveness. Surgical Assistant systems will work cooperatively
with human surgeons in carrying out precise and minimally invasive surgical
procedures.
This presentation will focus on the emerging role of medical robots
within CIS systems, with special attention to the synergy between the development
of image-guided, robotically-assisted delivery systems and the development
of novel minimally invasive localized therapies. It will draw upon
current and ongoing research in the newly established NSF Engineering Research
Center for Computer-Integrated Surgical Systems and Technology and elsewhere
to illustrate these themes.
Speaker Biography:
