|
 |
The autonomous motion primitives in the title,
"Development of Brain-Informatics Machines through Dynamical Connection
of Autonomous Motion Primitives," imply the control subsystems designed
to generate behaviors using sensory information. We first provide a humanoid
robot with engineered and designed autonomous motion primitives instead
of the synergetic structure that is innate in the sensory-motor system or
acquired using plasticity in the neural system of the humans. We took the
constructivist's approaches to structure the dynamics that determines transitions
among autonomous motion primitives, being motivated by the research paradigm
that various characteristics of the information processing of the brain
are the results of nonlinear dynamics of neural system.
The research team consisted of the following six research groups:
|
(Nakamura Group) |
PI: Prof. Yoshihiko Nakamura
Graduate School of Information Science and Technology
University of Tokyo |
|
| |
(Sasaki Group) |
PI: Prof. Masato
Sasaki
Graduate School of Education, University of Tokyo |
|
| |
(Yoshizawa Group) |
PI: Prof. Shuji Yoshizawa
College of Engineering
Saitama University |
|
| |
(Tsuchiya Group) |
PI: Kazuo Tsuchiya
Graduate School of Engineering
Kyoto University |
|
| |
(Asada Group) |
PI: Prof. Minoru Asada
Graduate School of Engineering
Osaka University |
|
| |
(Ushio Group) |
PI: Prof. Toshimitsu Ushio
Graduate School of Engineering Science
Osaka University |
| The outcomes of research can be categorized into the following four: |
|
A) |
Dynamics-Based Information Processing for Motion Recognition and Generation, |
|
| |
B) |
Mechanisms and Control for Further Evolution of Humanoids, |
|
| |
C) |
Measurements and Computation of Behaviors and Sensory Information, |
|
| |
D) |
Description and Information Processing of Motion Learning and Acquisition. |
| The main results of the research groups are listed with categorization as follows: |
|
A) |
Dynamics-Based Information
Processing for Motion Recognition and Generation |
|
(1) |
Dynamics-Based Information Processing for Generation
and Transition of Motion Patterns (Nakamura Group) |
| |
(2) |
Statistic Information Processing that Unifies Recognition
and Generation of Motion Patterns (Nakamura Group) |
| |
(3) |
Lyapunov Design of Dynamics-Based Information Processing
and Transition Control of Motion Patterns (Ushio Group) |
|
|
B) |
Mechanisms and Control for
Further Evolution of Humanoids |
|
(4) |
Humanoid Robot Design for Mobility Evolution (Nakamura
Group) |
| |
(5) |
Legged Walk Control with Nonlinear Oscillators that
Permit Discontinuous Phase Shifts (Tsuchiya Group) |
|
|
C) |
Measurements and Computation
of Behaviors and Sensory Information |
|
(6) |
Dynamics Computation and Behavior Measurements of Human
Figures (Nakamura Group) |
| |
(7) |
Visual Recognition Model and On-Line Linear Discrimination
(Yoshizawa Group) |
|
|
D) |
Description and Information
Processing of Motion Learning and Acquisition |
|
(8) |
Information Processing of Learning and Acquisition of
Behaviors (Asada Group) |
| |
(9) |
Ecological Description of Acts (Sasaki Group) |
|
|
|
|
