|
 Jun
Ueda
Assistant Professor
Education
-
Ph.D., Kyoto University, Japan, 2002
-
M.S., Kyoto University, Japan, 1996
- B.S., Kyoto University, Japan, 1994
Background
Dr. Ueda begins at Georgia Tech in May 2008 as Assistant Professor. Before
Georgia Tech, he was a Visiting Scholar and Lecturer at MIT, where he worked
on the development and control of cellular actuators inspired by biological
muscle. He developed compliant, large strain piezoelectric actuators and
a robust control method called stochastic broadcast feedback.
From 2002-2008 he was Assistant Professor at Nara Institute of Science and
Technology in Japan, where he led a research group dedicated to dynamics
and control in robotics, such as robot hand manipulation, tactile sensing,
and power-assisting.
From 1996 to 2202 and prior to obtaining his Ph.D, he worked at the Advanced
Technology R&D Center of Mitsubishi Electric Corporation in Japan. Here
he was involved in a variety of activities including disk drives, machine
tools, and satellite tracking antennas. His Ph.D. work at Kyoto University
was on the end-point control of a robot manipulator mounted on a nonrigid
base. He studied feedback control robustness in terms of the coupling of
the arm and base dynamics.
Research Areas
-
Automation and Mechatronics: Robotics (bio-inspired robot architecture,
actuators, sensors) and Automatic Control(Robust feedback control,
robustness, and flexible mechanism control)
Research
Robotics is an inherently interdisciplinary endeavor, requiring a deep
understanding of classical subjects and broad knowledge of emerging
scientific fields. 
Dr. Ueda's interest is in the coordinated design of robotic components,
such as mechanical and electrical. This coordinated design is the key
for new robotics as well as for practical mechatronics in industry.
Dr. Ueda's research goal is to establish a theoretical design framework
for highly integrated robotic systems. The application ranges from
industrial robots, medical and rehabilitation robots, to intelligent
assistive robots. Robustness against uncertainty of model and environment
is crucial for these robots which work closely with humans or physically
interact with humans and environments. The successful integration relies
on the coordinated design of control, structure, actuators, and sensors
by considering the dynamic interaction among them.
Actuators are a key component for new robotics. A cellular actuator concept
inspired by biological muscle structure connects many small actuator
units in series or in parallel, and composes in totality a single actuator.
The method of stochastic broadcast control coordinates a vast number
of units in this actuator system, which drastically improves wiring
and addressing issues. Piezoelectric actuators using nested architecture
with exponential strain amplification are developed for over twenty
percent strain that is comparable to natural skeletal muscles.
Dr. Ueda’s research also deals with novel robotic manipulation systems
and haptic devices. Robust control stability is investigated in the
presence of model uncertainty, redundancy, mechanical flexibility,
and nonlinearity.
Dynamic interaction between robot actuators and human muscles needs to
be considered for the development of intelligent prosthetics. A new
concept is "pinpointed" control of muscles that enables
modification of the load of selected muscles by exerting forces from
robot actuators. An integrated model of a musculoskeletal human model
and power assisting robot is developed and used for this control.
The combination of micro structure and stochastic control exhibits many
interesting aspects; even thermal fluctuation could be used as a random
signal source for the stochastic recruitment law. Technical challenges
include actuator-circuit integration by MEMS technology, wireless power
supply for actuator units, as well as theoretical analysis of the stability
and performance. Implementation of the cellular actuators to practical
mechanisms is also challenging, which would range from industrial application
such as micropositioner to entertainment/home application such as humanoids.
In the long term, he is interested in establishing a theoretical framework
that gives the abilities and limitations both at component level and
system level.
Distinctions
- IEEE/ASME Transaction on Mechatronics Technical Editor, 2008-present
- FANUC FA and Robot Foundation Original Paper Award, 2005
- Japan Society for the Promotion of Science postdoctoral fellowship for
research abroad, 2004
- Journal of the Robotics Society of Japan Editorial Board, 2004-2005
Representative Publications
-
Masahiro Kondo, Jun Ueda, and Tsukasa Ogasawara. 2008. Recognition of
In-Hand Manipulation using Contact State Transition for Multifingered
Robot Hand Control. Journal of Robotics and Autonomous
Systems 56(1),66-81.
Jun Ueda, Lael Odhner, and H. Harry Asada. 2007. Broadcast Feedback of
Stochastic Cellular Actuators Inspired by Biological Muscle Control.
The International Journal of Robotics Research 26(11-12), 1251--1265.
Jun Ueda, Atsutoshi Ikeda, and Tsukasa Ogasawara. 2005. Grip-Force Control
of an Elastic Object by Vision-Based Slip Margin Feedback During the
Incipient Slip. IEEE Transactions on Robotics 21(6), 1139-1147.
Jun Ueda and Tsuneo Yoshikawa. 2004. Robust Arm Configuration of Manipulator
Mounted on Flexible Base. IEEE Transactions on
Robotics 20(4), 781-
789.
Jun Ueda and Tsuneo Yoshikawa. 2004. Force Reflecting Bilateral Teleoperation
with Time Delay By Signal Filtering. IEEE Transactions
on Robotics and Automation 20(3), 613-619.
|
