Misc - NAKAMURA Takayuki

倒立二輪型移動ロボットによる人間の動的動作模倣のための身体部位マッピング

高橋佐多弥,高橋泰岳,前田陽一郎,中村恭之

第25回人工知能学会全国大会 ( 人工知能学会 )    2011.05

3次元距離情報と色情報の統合に基づく3次元物体追跡

中村恭之

ロボティクスメカトロニクス講演会2011 ( 日本機械学会 )    2A1-L07   2011.05

Person Tracking System with Active Camera

LI Peng, QI Yiqiang, WU Haiyuan, NAKAMURA Takayuki, MIURA Hirokazu, TAKI Hirokazu

電気学会研究会資料. IIS, 次世代産業システム研究会 = The papers of Technical Meeting on Innovative Industrial System, IEEE Japan   2011 ( 21 ) 23 - 26   2011.03

Detection of Grasping Position for Robot Hand Using Three-dimensional Vision without Shape Models

HIRATA Masaya, MIURA Hirokazu, NAKAMURA Takayuki, WU Haiyuan, TOMITA Fumiaki, NISHI Takao, MATSUDA Noriyuki, TAKI Hirokazu

電気学会研究会資料. IIS, 次世代産業システム研究会 = The papers of Technical Meeting on Innovative Industrial System, IEEE Japan   2011 ( 21 ) 15 - 18   2011.03

Classifier acceleration by imitation

Takahiro Ota, Toshikazu Wada, Takayuki Nakamura

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)   6495 ( 4 ) 653 - 664   2011  [Refereed]

　View Summary

This paper presents a framework named "Classifier Molding" that imitates arbitrary classifiers by linear regression trees so as to accelerate classification speed. This framework requires an accurate (but slow) classifier and large amount of training data. As an example of accurate classifier, we used the Compound Similarity Method (CSM) for Industrial Ink Jet Printer (IIJP) character recognition problem. The input-output relationship of trained CSM is imitated by a linear regression tree by providing a large amount of training data. For generating the training data, we developed a character pattern fluctuation method simulating the IIJP printing process. The learnt linear regression tree can be used as an accelerated classifier. Based on this classifier, we also developed Classification based Character Segmentation (CCS) method, which extracts character patterns from an image so as to maximize the total classification scores. Through extensive experiments, we confirmed that imitated classifiers are 1500 times faster than the original classifier without dropping the recognition rate and CCS method greatly corrects the segmentation errors of bottom-up segmentation method. © 2011 Springer-Verlag Berlin Heidelberg.

DOI

Development of body mapping from human demonstrator to inverted-pendulum mobile robot for imitation

Sataya Takahashi, Yasutake Takahashi, Yoichiro Maeda, Takayuki Nakamura

IEEE International Conference on Fuzzy Systems ( IEEE )    1344 - 1349   2011  [Refereed]

　View Summary

This paper addresses development of body mapping from a human demonstrator to an inverted-pendulum mobile robot for imitation. An inverted-pendulum mobile robot with torso and body links learns a dynamic kicking motion shown by a human. The robot observes the human demonstration with a camera, extracts the human region in each of images, maps the region to its own two links, estimates the link posture trajectories, and starts kicking motion learning based on the trajectory parameters for imitation. The mapping parameter gives an important role for successive imitation. A reasonable and feasible procedure of learning from observation for an inverted-pendulum robot and development of the body mapping is proposed and investigated in this paper. © 2011 IEEE.

DOI

Real-time 3-D object tracking using Kinect sensor

Takayuki Nakamura （Part： Lead author )

2011 IEEE International Conference on Robotics and Biomimetics, ROBIO 2011 ( IEEE )    784 - 788   2011  [Refereed]

　View Summary

In this paper, we propose a novel method for real-time three-dimensional (3-D) object tracking based on integration of 3-D range and color information. To get 3D range and color information simultaneously, our method utilizes the Microsoft Kinect sensor which is composed of depth and color cameras. 3-D range and color information are integrated using the estimated depth and color cameras intrinsic parameters and relative transformation between the cameras. The target region can be tracked by processing depth pixels with color information. Through experiment in real environment, we confirm the validity of our method. © 2011 IEEE.

DOI

Inverted-Pendulum Mobile Robot Motion Learning from Human Player Observation

Sataya Takahashi, Hiroaki Nonoshita, Yasutake Takahashi, Y.Maeda, Takayuki Nakamura

Proceedings of Joint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems ( Japan Society for Fuzzy Theory and intelligent informatics (SOFT) )    pp.211-216   2010.12  [Refereed]

方策勾配法を用いた倒立振子型移動ロボットの蹴球動作学習

野々下博昭,高橋泰岳,中村恭之,前田陽一郎

第28回日本ロボット学会学術講演会 ( 日本ロボット学会 )    2010.09

Behavioral Development of Ball Kicking Motion of a Two-wheeled Inverted Pendulum Mobile Robot

Yasutake Takahashi, Hiroaki Nonoshita, Takayuki Nakamura, Yoichiro Maeda

2010 IEEE INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS (FUZZ-IEEE 2010) ( IEEE )    pp.830-835   2010  [Refereed]

　View Summary

In this paper, we introduce a method for generating a dynamic motion such that a two-wheeled inverted pendulum robot kicks a ball far away utilizing its own body dynamics while it keeps standing. Such a dynamic motion can be acquired through trial and error based on a reinforcement learning scheme. We utilize a simple policy gradient method to acquire a kicking motion which is designed by defining the desired parameters such as body angle, wheel angular velocity and so on. To show the validity of our approach, we perform computer simulation experiments of behavior acquisition for the two-wheeled inverted pendulum robot. Based our approach, we succeeded in acquiring the kicking motion of the two-wheeled inverted pendulum robot. A very interesting finding is: each of the acquired motions deviates from the desired trajectory, which is given by the human designer while keeping evaluation value of the acquired motion high.

2P1-F06 Picking Up By A Wheeled Inverted Pendulum Manipulator

Nakamura Takayuki

ロボティクス・メカトロニクス講演会講演概要集 ( The Japan Society of Mechanical Engineers )  2009   "2P1 - F06(1)"-"2P1-F06(3)"   2009.05

　View Summary

In this paper, we propose a new nonlinear programming based motion generator for a wheeled inverted pendulum manipulator. Our method utilizes L-BFGS-B method which can handle inequality constraints intrinsically and whose computational cost is low. Our method can generate various motions of the robot only by setting up an objective function for a given task. Through computer simulation, we confirm the validity of our method.

2A2-D18 Learning Dynamic Throwing Motion of A Wheeled Inverted Pendulum utilizing Whole Body Dynamics

Nishikawa Takehiro, TAKAHASHI Yasutake, NAKAMURA Takayuki, ASADA Minoru, ISHIGURO Hiroshi

ロボティクス・メカトロニクス講演会講演概要集 ( The Japan Society of Mechanical Engineers )  2009   "2A2 - D18(1)"-"2A2-D18(4)"   2009.05

　View Summary

We apply reinforcement learning to a wheeled inverted pendulum robot that acquires dynamic throwing motion utilizing whole body dynamics. Large number of parameters are needed to be calibrated so that the robot becomes able to throw a ball far away utilizing its own body dynamics while it keeps standing. We investigated the learning process of the throwing motion by application of a policy gradient method with a dynamics simulater.

2P2-C12 A Coarse-to-Fine Strategy for Fast Scan Matching

Nakamura Takayuki, Maeda Yukihiro

ロボティクス・メカトロニクス講演会講演概要集 ( The Japan Society of Mechanical Engineers )  2008   "2P2 - C12(1)"-"2P2-C12(4)"   2008.06

　View Summary

In this paper, we propose a fast and reliable scan matching method so that it can be used in the cluttered environment. To speed up the processing time, our method utilize coarse-to-fine strategy. First, our method employs histgram matching approach to estimate displacements of position and orientation roughly. Second, to estimate them accurately, our method employ constrained least square method based on the formulation using poin-to-line metric. Such metric is also effective to the cluttered environment. The experimental results show the usefulness of our method.

2P2-H02 Generating Various Motions of Multiple Articulated Mobile Robot with Nonlinear Programming

Nakamura Takayuki

ロボティクス・メカトロニクス講演会講演概要集 ( The Japan Society of Mechanical Engineers )  2008   "2P2 - H02(1)"-"2P2-H02(4)"   2008.06

　View Summary

In this paper, we propose a new nonlinear programming based inverse kinematics solver for generating various motions of a multiple articulated robot. Our method utilizes L-BFGS-B method which can handle inequality constraints intrinsically and whose computational cost is low. Our method can generate various motions of a multiple articulated robot only by setting up objective functions. In this paper, we provide several objective functions one of which can generate the motion of a floating-base robot and which can generate self-motion of the redundant robot without calculating the pseudoinve...

Generating Various Motions of Multiple Articulated Mobile Robot with Nonlinear Programming

The 10th International Conference on the SIMULATION OF ADAPTIVE BEHAVIOR (SAB'08)   P28   2008  [Refereed]

測域センサデータの直線追跡に基づくスキャンマッチング法

武野哲也, 中村恭之, 和田俊和

画像の認識・理解シンポジウム(MIRU2007)     1564 - 1569   2007.07  [Refereed]

主成分木を用いた写像学習法の提案と性能比較

荒井英剛,中村恭之,和田俊和

MIRU2007画像の認識・理解シンポジウム論文集     pp. 516--521   2007.07

2P1-G05 Local Environmental Map Generation Based on Tracking Lines in Laser Range Data

Takeno Tetsuya, Nakamura Takayuki, Wada Toshikazu

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 )  2007   "2P1 - G05(1)"-"2P1-G05(3)"   2007.05

　View Summary

We propose a new scan matching algorithm that makes use of the simple polygonal structure of environment. Our method extracts line segments from a current scan and matches them with line segments in a previous scan, that is, tracks the line segments in the sequence of scan. Our method simultaneously computes the pose shift (Δx ,Δy,Δθ) based on the correspondence between line segments in the two successive scans and the transformation such that the current scan is mapped optimally to the previous scan. By the use of line segments in the scan, it is possible to detect invariants for position ...

2A2-F07 Self Localization Method for Vision-based Mobile Robot Using Multiple Parallel Line Patterns

Kawano Hiroaki, Nakamura Takyuki, Yokomori Ryosuke, Chen Qian, Wada Toshikazu

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 )  2007   "2A2 - F07(1)"-"2A2-F07(3)"   2007.05

　View Summary

We propose a new self-localization method for a mobile robot equipped with a fixed-viewpoint active camera by observing multiple parallel lines patterns with the camera. The method utilizes the technique for estimating location and posture of the camera from one parallel lines pattern which was previously developed by our research group. The method switches from a certain parallel lines pattern used for self-localization to another parallel lines pattern while the robot is moving in the environment. After the new parallel lines pattern is selected, the method keeps estimating location and p...

分離度フィルタと PaLM-treeを用いた視線方向の推定

松本拓也, 呉海元, 和田俊和, 中村恭之

画像の認識・理解シンポジウム(MIRU2006)     1042 - 1047   2006.07  [Refereed]