著者
日下 航 尾形 哲也 小嶋 秀樹 高橋 徹 奥乃 博
出版者
一般社団法人 日本ロボット学会
雑誌
日本ロボット学会誌 (ISSN:02891824)
巻号頁・発行日
vol.28, no.4, pp.532-543, 2010 (Released:2012-01-25)
参考文献数
29
被引用文献数
1

We propose a model of evolutionary communication with voice signs and motion signs between two robots. In our model, a robot recognizes other's action through reflecting its self body dynamics by a Multiple Timescale Recurrent Neural Network (MTRNN). Then the robot interprets the action as a sign by its own hierarchical Neural Network (NN). Each of them modifies their interpretation of signs by re-training the NN to adapt the other's interpretation throughout interaction between them. As a result of the experiment, we found that the communication kept evolving through repeating miscommunication and re-adaptation alternately, and induced the emergence of diverse new signs that depend on the robots' body dynamics through the generalization capability of MTRNN.
著者
細田 耕 浅田 稔
出版者
The Robotics Society of Japan
雑誌
日本ロボット学会誌 (ISSN:02891824)
巻号頁・発行日
vol.14, no.2, pp.313-319, 1996-03-15 (Released:2010-08-10)
参考文献数
12
被引用文献数
3 4

This paper describes an adaptive visual servoing controller consisting of an on-line estimator and a feedback/feedforward controller for uncalibrated camera-manipulator systems. The estimator does not need a priori knowledge on the kinematic structure nor parameters of the camera-manipulator system, such as camera and link parameters. The controller consists of feedforward and feedback terms to make the image features converge to the desired trajectories, by using the estimated results. Some experimental results demonstrate the validity of the proposed estimator and controller.
著者
長谷川 昂宏 山内 悠嗣 山下 隆義 藤吉 弘亘 秋月 秀一 橋本 学 堂前 幸康 川西 亮輔
出版者
一般社団法人 日本ロボット学会
雑誌
日本ロボット学会誌 (ISSN:02891824)
巻号頁・発行日
vol.36, no.5, pp.349-359, 2018 (Released:2018-07-15)
参考文献数
27

Automatization for the picking and placing of a variety of objects stored on shelves is a challenging problem for robotic picking systems in distribution warehouses. Here, object recognition using image processing is especially effective at picking and placing a variety of objects. In this study, we propose an efficient method of object recognition based on object grasping position for picking robots. We use a convolutional neural network (CNN) that can achieve highly accurate object recognition. In typical CNN methods for object recognition, objects are recognized by using an image containing picking targets from which object regions suitable for grasping can be detected. However, these methods increase the computational cost because a large number of weight filters are convoluted with the whole image. The proposed method detects all graspable positions from an image as a first step. In the next step, it classifies an optimal grasping position by feeding an image of the local region at the grasping point to the CNN. By recognizing the grasping positions of the objects first, the computational cost is reduced because of the fewer convolutions of the CNN. Experimental results confirmed that the method can achieve highly accurate object recognition while decreasing the computational cost.
著者
白井 達也 金子 真 原田 研介 辻 敏夫
出版者
The Robotics Society of Japan
雑誌
日本ロボット学会誌 (ISSN:02891824)
巻号頁・発行日
vol.20, no.1, pp.57-64, 2002-01-15 (Released:2010-08-25)
参考文献数
21
被引用文献数
1

Through grasp experiments by human achieving an enveloping grasp for a small cylindrical object placed on a table, we found an interesting grasping motion, where human changes the finger posture from upright to curved ones after each finger makes contact with the object. During this motion, the object is automatically lifted up through either rolling or sliding motion between the finger tip and the object. A series of this motion is called as Detaching Assist Motion (DAM) . An advantage of DAM is that most of grasping motions can be done on the table instead of in the air. Therefore, we can avoid the worst scenario where the object falls down to the table. We first discuss the basic mechanism of DAM by human experiments. We then apply the DAM to a grasping motion by a multi-fingered robot hand. We show that the DAM can be explained by using Self-Posture Changing Motion. We also show some simulation and experimental results to confirm that a small object can be grasped easily by applying the DAM.
著者
赤井 孝幸
出版者
日本ロボット学会
雑誌
日本ロボット学会誌 (ISSN:02891824)
巻号頁・発行日
vol.15, no.1, pp.28-31, 1997-01-15 (Released:2010-08-10)
参考文献数
4
著者
Takashi Katsuragawa Kiyoshi Ioi Nobuyuki Kubota Osamu Noro
出版者
The Robotics Society of Japan
雑誌
Journal of the Robotics Society of Japan (ISSN:02891824)
巻号頁・発行日
vol.12, no.6, pp.893-898, 1994-09-15 (Released:2010-08-25)
参考文献数
9
被引用文献数
3 11

In order to execute various contacting tasks, a robot must be controlled contacting force or contacting resistance. In the former case, a force-control is available, and in the latter case, an impedance-control and a compliance-control are useful.In this paper, we propose a Hybrid Compliance/Force Control (H. C. C.) which is a combination of a Hybrid Position/Force Control and a compliance-control. In this method, a position control, a force-control, and a compliance-control can be switched for any direction of a given task. Moreover we compose typical 2 types of H. C. C. (torque-based and position-based), and both methods are installed in a 6-D. 0. F. industrial robot.This study shows compositions of 2 types of H. C. C. and their experimental results.
著者
清水 昌幸 角谷 啓 尹 祐根 北垣 高成 小菅 一弘
出版者
日本ロボット学会
雑誌
日本ロボット学会誌 (ISSN:02891824)
巻号頁・発行日
vol.25, no.4, pp.606-617, 2007-05-15 (Released:2010-08-25)
参考文献数
22
被引用文献数
3 1

This paper addresses the inverse kinematic problem for 7 DOF anthropomorphic manipulators with joint limits. Specifically, the paper analyzes how the joint limits restrict the redundancy to develop practical inverse kinematic resolutions under the joint limits. First, the manipulator model considered in this paper is provided. Second, how to solve the inverse kinematics of the redundant manipulator is described. Third, how joint limits affect the inverse kinematic solutions is investigated. Forth, several suggestions for resolving redundancy under the joint limits are presented. Lastly, the effectiveness of the redundancy resolutions is illustrated by simulation results.
著者
遠山 茂樹 村久木 康夫
出版者
The Robotics Society of Japan
雑誌
日本ロボット学会誌 (ISSN:02891824)
巻号頁・発行日
vol.15, no.5, pp.703-709, 1997-07-15
参考文献数
8
被引用文献数
1

近年, 産業界では工場の自動化に伴い定置式の産業用ロボットに代わる移動機能を備えた自律移動ロボットや無人搬送車などへの要望が強くなってきている. これらは, その移動機構により車輪式, クローラ式, 特殊機構などの3種類に分けられるが, なかでも車輪式のものが他の機構と比べてエネルギー効率が良く, 機構が簡単であることから最もよく利用されている. さらに車輪式自律移動ロボットには, 2駆動1キャスタ (2DWlC), 1駆動1操舵 (1DWlS), 全方向移動車などがある.<br>このような車輪式自律移動車は, タイヤと路面との摩擦力で運動を行っている. このためタイヤと路面の間には必ず滑りが生じ, その運動を不確かなものにしている. 車輪式自律移動車に関する研究においては, 位置決め制御や軌道追従制御などの研究結果が数多く発表されてきたが [1]-[3], タイヤの滑りや移動機構を含めた動力学的なシミュレーションを行い, それら評価を行った研究結果はほとんど発表されていない. また, その移動機構の設計においては, 経験的に設計されたものがほとんどである.<br>本研究の目的は, 移動車両の動力学的なシミュレーションを行うことによって, 移動機構の構造やタイヤの性質が, 車両の運動性能に及ぼす影響を明らかにするとともに, 軌道追従制御の評価を行い, その結果を最適設計に役立てることである. その一環として本研究では, 当研究室で開発を行ってきた汎用の機構の運動解析シミュレータA1 MOTIONに, タイヤの力学に基づいたタイヤの力要素およびタイヤを駆動するDCモータの力要素, 減速機をモデル化したものを組み込み, 車輪式自律移動車の走行シミュレーションを行うことができるシミュレータの開発を行った.<br>そして, 本シミュレータの有用性を検証するために, 前輪キャスタ方式のマウス (小型自律移動車) の走行実験を行い, シミュレーションによる走行結果との比較を行った. その結果, このシミュレータはマウスの運動をうまく再現できるという結論を得ることができた. そして, 本シミュレータを用いて, タイヤの剛性, ホイールベース, 車軸幅, キャスタ半径などがその運動性能にどのような影響を与えるのかについて明らかにし, 軌道追従制御の評価を行った.
著者
上田 隆一 新井 民夫
出版者
日本ロボット学会
雑誌
日本ロボット学会誌 (ISSN:02891824)
巻号頁・発行日
vol.25, no.1, pp.103-112, 2007-01-15 (Released:2010-08-25)
参考文献数
25
被引用文献数
2 1

We propose the real-time Q-MDP value method for decision making of a robot under uncertain state recognition. When the computation result of a control problem is known on the assumption that recognition is certain, the original Q-MDP value method decides an appropriate action based on uncertain recognition. The method is not suitable for real-time decision making due to the complexity of probability calculation. In the real-time Q-MDP value method, a particle filter that is utilized for state estimation is directly used for the probability calculation. The proposed method can make it possible to execute the Q-MDP value method in real-time. The proposed method is applied to total behavior of a goalkeeper for robot soccer competition. Experiments and actual games have suggested that this method can decide actions effectively according as uncertain result of state estimation.
著者
大須賀 公一 吉田 浩治 小野 敏郎
出版者
日本ロボット学会
雑誌
日本ロボット学会誌 (ISSN:02891824)
巻号頁・発行日
vol.14, no.7, pp.1018-1025, 1996-10-15 (Released:2010-08-25)
参考文献数
9
被引用文献数
2 6

In this paper, we propose a new design concept of manipulator which is called “Torque-Unit Manipulator (TUM) ”. The Torque-Unit Manipulator has an open-loop kinematic chain as the typical traditional manipulators do, however each joint of which has one degree of freedom free joint. An actuator called “torque unit” is mounted on an arbitrary place of the link. The merit of the TUM would be an easy maintenance. In this paper, the dynamic model and the position-controllability of the N degree-of-freedom general TUM are shown in detail. A control system is designed for the TUM and its effectiveness is shown by a simple simulation.
著者
志村 康治 堀 洋一
出版者
The Robotics Society of Japan
雑誌
日本ロボット学会誌 (ISSN:02891824)
巻号頁・発行日
vol.11, no.2, pp.235-245, 1993-03-15
参考文献数
17
被引用文献数
8 2

It is important that we control a robot manipulator considering contact with an object. In force control we have two difficulties caused by (1) environmental parameter change and (2) large contact force at collision. The collision always happens in a short transient period from trajectory control to force control. In this paper, we propose a robustifid force and collision control method. First, we designed TDOF (Two Degrees Of Freedom) force controller. This force controller enables us to obtain stable and fast response in the face of considerable environmental parameter variation. Next, in order to reduce the shock and also to change control modes smoothly from position to force, we proposed a novel collision control method by using feedforward position command together with the robust force controller based on position. We realized these proposals by using a TRANSPUTER based con-troller and confirmed their usefulness with experiments by using a 1-axis and a lifesize 6-axis mani-pulator.