Misc - NAKAMURA Takayuki

Fast Self-Localization Method for Mobile Robots Using Multiple Ominidirectional Vision Sensors

NAKAMURA Takayuki, OHHARA Masahichi, OGASAWARA Tsukasa, ISHIGURO Hiroshi

IPSJ SIG Notes. CVIM ( 一般社団法人情報処理学会 )  2001 ( 4 ) 133 - 138   2001.01

　View Summary

In the multi-agent environment, it is important to identify position and orientation of multiple robots for accomplishing a given task in cooperative manner. This paper proposes a method for estimating position and orientation of multiple robots using multiple omnidirectional images based on geometrical constraints. Our method reconstruct not only relative configuration between robots but also absolute one usign the knowledge of landmarks in the environment. Even if there are some obstacles in the environment, our method can estimate relative configuration between robots based on the result...

Continuous valued Q-learning method able to incrementally refine state space

M Takeda, T Nakamura, T Ogasawara

IROS 2001: PROCEEDINGS OF THE 2001 IEEE/RJS INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS, VOLS 1-4 ( IEEE )    265 - 271   2001  [Refereed]

　View Summary

The conventional reinforcement learning method has trouble in applying to real robot tasks, because such method must be able to represent the values in terms of infinitely many state and action pairs. In order to represent a action value function continuously, a function approximation method is usually applied. In our previous work [1], we pointed out that this type of learning method potentially has a discontinuity problem of optimal actions given a state.
In this paper, we propose a method for estimating where discontinuity of optimal action takes place and for refining a state space incrementally. We call this method an continuous valued Q-learning (hereafter, CVQ-learning) method. To show the validity of our method, we apply the method to a simulated robot.

Towards Cognitive Agents: Embodiement based Object Recognition for Vision-Based Mobile Agents

Kazunori Terada, Takayuki Nakamura, Hideaki Terada, Tsukasa Ogasawara

Prof. of 2000 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS'2000)     2067-2072   2000.11  [Refereed]

対象認識のための接触に基づく効用関数の生成 (小特集 身体性)

寺田 和憲, 中村 恭之, 小笠原 司

人工知能基礎論研究会 ( 人工知能学会 )  42   45 - 50   2000.09

複数の全方位画像を用いた実時間自己位置同定

大原 正満、海老名 明弘、中村 恭之、今井 正和、小笠原 司、石黒浩

第18回ロボット学会学術講演会予稿集     1225 - 1226   2000.09

Embodiment based Object Recognition

TERADA Kazunori, NAKAMURA Takayuki, TAKEDA Hideaki, OGASAWARA Tsukasa

Proceedings of the Annual Conference of JSAI ( 人工知能学会 )  14   508 - 511   2000.07

　View Summary

In order to keep visual tracking systems with color segmentation technique running in real environment, it should be developed on-line learning method to update models for adapting them to dynamic changes of surroundings.<BR>To deal with this problem, we propose an on-line visual learning method for color image segmentation and object tracking in dynamic environment. Our method utilizes Fuzzy ART architecture which is a kind of neural network for competitive learning. The mechanism of this neural network is suitable for on-line learning and different from that of backpropagation type neural...

複数の全方位画像を用いた実時間自己位置同定

海老名 明弘, 中村 恭之, 今井 正和, 小笠原 司

ロボティクス・メカトロニクス講演会講演概要集 ( 一般社団法人日本機械学会 )  2A1-67-090   72 - 72   2000.05

　View Summary

複数の移動ロボットに搭載した全方位カメラを用いた実時間自己位置同定手法について述べる。各移動ロボットにおいて得られた全方位画像を床面上への画像に変換し, 線分を検出する。検出された線分とゴールの方向や位置情報を用いて定量的な自己位置同定を行う。複数のロボット間でこれらの情報を交換する事で, 各ロボットの自己位置やロボット間の相対的な位置関係を実時間で同定した。サッカーロボットを用いて行った実験結果について述べると共に, 問題点と今後の課題について述べる。

最適行動の不連続問題を考慮したContinuous Valued Q-learningの提案

武田政宣,中村恭之,今井正和,小笠原司,浅田稔

第5回ロボティクスシンポジア予稿     409 - 414   2000.03  [Refereed]

身体性に基づく環境モデリングと行動生成

寺田和憲,中村恭之,武田英明,小笠原司

第5回ロボティクスシンポジア予稿集     92 - 97   2000.03  [Refereed]

Toward a Visual Tracking System with On-Line Visual Learning Capability

NAKAMURA T.

Journal of Robotics Society of Japan   18 ( 7 ) 1047 - 1054   2000

Enhanced continuous valued Q-learning for real autonomous robots

M Takeda, T Nakamura, M Imai, T Ogasawara, M Asada

Proc. of Int. Conf. of The Society for Adaptive Behavior 2000 ( VSP BV )    195 - 202   2000  [Refereed]

DOI

Real-time estimating spatial configuration between multiple robots by triangle and enumeration constraints

T. Nakamura, A. Ebina, M. Imai, T. Ogasawara, H. Ishiguro

IEEE International Conference on Intelligent Robots and Systems   3   2048 - 2054   2000  [Refereed]

　View Summary

In the multi-agent environment, it is important to identify position and orientation of multiple robots for accomplishing a given task in cooperative manner. This paper proposes a method for estimating position and orientation of multiple robots using multiple omnidirectional images based on geometrical constraints. Our method reconstruct not only relative configuration between robots but also absolute one using the knowledge of landmarks in the environment. Even if there are some obstacles in the environment, our method can estimate absolute configuration between robots based on the results of self-localization of each robot.

DOI

ロボットサッカー研究のためのシミレーター開発

藤原広光,中村恭之,今井正和,小笠原司

第17回日本ロボット学会学術講演会予稿集     255 - 256   1999.09

RoboCup '98パリ大会

中村 恭之

Journal of the Robotics Society of Japan ( The Robotics Society of Japan )  17 ( 5 ) 643 - 644   1999.07

DOI

A Nonholonomic Mobile Robot Navigation in Unknown Environment

FUJII Hirokazu, HIRANO Masatoshi, NAKAMURA Takayuki, MATSUMOTO Yoshio, IMAI Masakazu, OGASAWARA Tsukasa

電気学会研究会資料. SC, システム制御研究会   1999 ( 11 ) 35 - 40   1999.03

Mobile Robot Navigation in a Dynamic Environment using KL Expansion of Omnidirectional Images

SHIBATA Atsushi, NAKAMURA Takayuki, MATSUMOTO Yoshio, IMAI Masakazu, OGASAWARA Tsukasa

電気学会研究会資料. SC, システム制御研究会   1999 ( 11 ) 41 - 46   1999.03

色情報に基づくラベリング処理LSIの開発とその応用

秋田純一,渡辺晃,中村恭之,小笠原

第４回ロボティクスシンポジア予稿集     189 - 194   1999.03  [Refereed]

実時間競合学習モデルFuzzy ARTを用いたカラー画像の領域分割と追跡

中村恭之,小笠原司

第4回ロボティクスシンポジア     183 - 188   1999.03  [Refereed]

Why is cognitive robotics promising?

NAKAMURA Takayuki, SATO Tomomasa, KUNIYOSHI Yasuo, HIRAKI Kazuo, SHIBATA Tomohiro, ASADA Minoru, MACDORMAN Karl F, TANI Jun

Journal of the Robotics Society of Japan ( The Robotics Society of Japan )  17 ( 1 ) 38 - 43   1999.01

DOI

On-Line Visual Learning Method for Color Image Segmentation and Object Tracking

NAKAMURA T.

Proc. of IROS'99     222 - 228   1999  [Refereed]