日本ロボット学会誌
Online ISSN : 1884-7145
Print ISSN : 0289-1824
ISSN-L : 0289-1824
17 巻 , 1 号
選択された号の論文の26件中1~26を表示しています
  • 石黒 浩
    1999 年 17 巻 1 号 p. 1
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • 浅田 稔, 石黒 浩, 國吉 康夫
    1999 年 17 巻 1 号 p. 2-6
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • 銅谷 賢治
    1999 年 17 巻 1 号 p. 7-10
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • 認知モデリングにおける仮説形成ツールとしてのロボット
    開 一夫
    1999 年 17 巻 1 号 p. 11-15
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • 中島 秀之
    1999 年 17 巻 1 号 p. 16-19
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • マクドーマン ・カール
    1999 年 17 巻 1 号 p. 20-24
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • 谷 淳
    1999 年 17 巻 1 号 p. 25-28
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • 國吉 康夫, ベルトゥーズ ・リュク
    1999 年 17 巻 1 号 p. 29-33
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • 浅田 稔
    1999 年 17 巻 1 号 p. 34-37
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • 中村 恭之, 佐藤 知正, 國吉 康夫, 開 一夫, 柴田 智広, 浅田 稔, マクドーマン カール, 谷 淳
    1999 年 17 巻 1 号 p. 38-43
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • 佐藤 知正
    1999 年 17 巻 1 号 p. 44-45
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • 大八木 衆司, 脇坂 達也, 井上 文宏, 池田 雄一, 渡辺 幸次, 若林 明, 猫本 善続, 榎本 健次, 神吉 博
    1999 年 17 巻 1 号 p. 46-47
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • 橋本 学, 鷲見 和彦, 羽下 哲司, 黒田 伸一, 小平 紀生, 岩田 彰太郎
    1999 年 17 巻 1 号 p. 48-49
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • 駒田 諭
    1999 年 17 巻 1 号 p. 50
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
  • 湯 軍, 渡辺 桂吾, 栗林 勝利, 白石 大和
    1999 年 17 巻 1 号 p. 51-60
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
    In this paper, we describe a holonomic omnidirectional mobile robot with three orthogonal-wheel assemblies and its high accurate tracking control system. We first analyze three types in the realization of mechanical structure. We then develop an omnidirectional mobile robot using the lateral wheel assembly to obtain the much simpler control system. Several test results in the feedforward control system indicated that the mobile robot had a desirable full mobility and smooth motion. In order to realize a dynamic control system based on the model information, the dynamic model and its resolved acceleration control system are described. The effectiveness of the control system is demonstrated by a set of practical experiments.
  • 鄭 宇眞, 中村 仁彦
    1999 年 17 巻 1 号 p. 61-67
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
    Exploiting the unique features of nonholonomic systems, we have proposed and fabricated the nonholonomic manipulator which is a controllable n joint manipulator with only two actuators. In order to create its nonholonomic constraint, a special type of velocity transmission, called the nonholonomic gear, was used.
    There are many possible alternatives of designing underactuated manipulators using the nonholonomic gear. The nonholonomic manipulator was designed focusing on the mechanical simplicity. As a result, we have faced with several problems in building a manipulator with many number of joints. In this paper, we propose the chained form manipulator which is designed to solve mechanical and mathematical problems. Presented simulation results show that the new design satisfies control simplicity.
  • 平井 慎一
    1999 年 17 巻 1 号 p. 68-83
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
    A systematic approach to the kinematic and static analysis of manipulative operations performed through mechanical contacts is presented. A variety of manipulation problems, including assembly and grasps, have been treated separately in different streams of robotics research. All of the problems are treated as problems to solve a certain class of inequalities resulting from the unidirectional nature of mechanical contacts. One of the fundamental difficulties in the analysis of manipulative operations is the intractable nature of inequalities.
    In this paper, we establish an underpinning mathematical tool for dealing with a variety of manipulative operations that are governed by unidirectional constraints. First, we introduce a coherent representation for formulating various manipulation problems. Second, we develop several procedures based on the theory of polyhedral convex cones to solve these problems in a systematic and straightforward manner. The method is then implemented on a computer and is applied to a variety of manipulation problems including grasping, fixturing, assembly, and admittance control.
  • CPSアクティブタッチ融合型地図生成法
    倉爪 亮, 広瀬 茂男, 岩崎 倫三, 長田 茂美
    1999 年 17 巻 1 号 p. 84-90
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
    To realize the automatic creation of an environmental map using a mobile robot, i) accurate positioning of the mobile robot itself and ii) accurate and reliable measurements of the relative positions of the surroundings are requisite. However, conventional methods for i), such as dead-reckoning or the placement and measurement of several landmarks, have low accuracy of position measurement. In addition, some typical devices that are used for ii), such as CCD cameras, ultrasonic sensors, and laser range finders, are likely to produce inaccurate measurements depending on lighting conditions and problematic objects, such as mirrors. In this paper, a new method named “Map Creation by CPS Based Active Touch” that is able to create an environmental map with high accuracy and reliability is proposed. This method employs the “Cooperative Positioning System (CPS)” that we have proposed for mobile robot positioning and “active touch” that is able to gather highly reliable information from a robot which contacts its surroundings as it moves. Furthermore, we propose a bumper mechanism that can be utilized not only to absorb shock but also to realize purely mechanical wall following behavior. Experiments in an indoor environment show that the proposed method is able to create an accurate environmental map of the positions of major features such as walls, pillars and doors.
  • 実システムによる基本ブロック組積作業と精度
    見浪 護, 羽多野 正俊, 朝倉 俊行
    1999 年 17 巻 1 号 p. 91-100
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
    This research is concerned with a construction operation system to make a long fence with piling up blocks by a mobile manipulator. The long fence can not be built by a floor-fixed robot manipulator. Then a kind of moving mechanism is essential to the manipulator. A mobile manipulator having the moving mechanism has an ability to build a huge structure like the long fence, because it can move around in an operation area by itself. One of problems to pile up blocks accurately by the mobile manipulator is errors between a desired position/orientation at destination and an actual one of a mobile robot which is a body of the mobile manipulator. The errors disturb accurate construction operations. A purpose of this research is to build the mobile manipulator system to be able to construct the huge structure. In this paper, a mobile manipulator system having a recognition system to detect the position/orientation errors of the mobile manipulator is proposed. First, a hand-eye camera system is constructed to recognize the position/orientation errors caused by traveling. Second, a fixation control method of the hand-eye camera is utilized to recognize the errors precisely. Third, the real construction operation system with the above recognition system is constructed. Then, the accuracy of piled blocks is discussed by experiments using our real mobile manipulator system, as one of operations to construct the huge structure. Finally, influences caused by changes of yaw and pitch angle of the body of the mobile manipulator are evaluated with real experiments.
  • 辻尾 昇三, 中村 勇人
    1999 年 17 巻 1 号 p. 101-109
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
    Coordinated compliant motion of two robotic devices, i.e., a manipulator-positioner system, can introduce greater flexibility and dexterity into manufacturing and improve the productivity of their processes. This paper addresses the coordinated compliant motion control of a manipulator-positioner system without using a geometric model of a workpiece. Dynamics and control of manipulators under time-dependent holonomic constraints are studied by formulating coordinated compliant motion of manipulator-positioner systems as constrained motion of a manipulator under geometric end-effector constraints imposed by a positioner. A virtual constraint plane is introduced to prescribe a desired position trajectory of an end-effector of a manipulator. A compliant motion control scheme without using a geometric model of a workpiece is realized by fusing the measurements of the articular displacements and velocities of a manipulator and a positioner and the constraint force of an end-effector of a manipulator. A numerical simulation and an experiment for coordinated contour-following tasks are conducted to demonstrate the feasibility of the proposed compliant motion control scheme without using a workpiece geometric model.
  • 目標ダイナミクス法の実ロボットへの適用
    中西 淳, 福田 敏男, Daniel E. Koditschek
    1999 年 17 巻 1 号 p. 110-117
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
    We report on our empirical success in the study of two-link brachiating robot. In our previous work, we have proposed a new algorithm, the “target dynamics method.” Inspired by the pendulum-like motion of an ape's brachiation, the task is encoded as an output of an appropriately chosen target dynamical system. Numerical simulations indicate that the resulting controller solves a number of brachiation problems that we term the “ladder”, “swing up” and “rope” problems. In this paper, the proposed controller is experimentally implemeted on a two-link brachiating robot. The swing locomotion and swing up behavior of the robot as well as continuous locomotion have been successfully achieved. The experimental results illustrate the validity of our control strategy.
  • 高橋 泰岳, 浅田 稔
    1999 年 17 巻 1 号 p. 118-124
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
    Reinforcement learning has recently been receiving increased attention as a method for robot learning with little or no a priori knowledge and higher capability of reactive and adaptive behaviors. However, there are two major problems in applying it to real robot tasks: how to construct the state space, and how to accelerate the learning. This paper presents a method by which a robot learns a purposive behavior within less learning time by incrementally segmenting the sensor space based on the experiences of the robot. The incremental segmentation is performed by constructing local models in the state space, which is based on the function approximation in terms of the sensor outputs and the reinforcement signal to reduce the learning time. The method is applied to a soccer robot which tries to shoot a ball into a goal. The experiments with computer simulations and a real robot are shown. As a result, our real robot has learned a shooting behavior within less than one hour training by incrementally segmenting the state space.
  • 吉留 忠史, 兵頭 和人, 小林 博明
    1999 年 17 巻 1 号 p. 125-132
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
    A tendon-driven robotic mechanism to work with persons co-operatively in the same area is investigated. The mechanism is driven by tendons that are more than the number of degrees of freedom and have non-linear elasticity. It has the advantage of the joint compliance adjustability. This is realized by the tendon redundancy. On the other hand, it has the disadvantage of the breakage of tendons. This makes it difficult to assure the safety of persons working co-operatively with it. But the tendon redundancy also allow us to compensate this disadvantage.
    This paper describes a method to cope with the tendon breakage for a tendon-driven robotic mechanism. After the basic results for the mechanism are introduced briefly, the degree of safety is defined as the maximum number of tendons allowed to be broken simultaneously without losing the control. Some results are shown with respect to the degree of safety. These are useful to design a tendon-driven robotic mechanism. Next, a method to compensate for the loss of tendon forces of broken tendons with the other tendons is discussed. It allows us safety are broken down simultaneously. Finally, simulation results are given to show the effect of the method.
  • 原田 研介, 金子 真
    1999 年 17 巻 1 号 p. 133-139
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
    This paper discusses the robustness of the equilibrium grasp of multiple objects. We first provide a general mathematical formulation on the kinematic relationship for multiple objects grasped by a multi-fingered robot hand assuming that the rolling condition is satisfied at each contact point. We define the rolling based redundancy under gravitational field, in which the grasped objects have degrees of freedom of motion through rolling even if the finger posture is fixed. When the finger-object system is rotated, the equilibrium before rotation is lost if the system has the rolling based redundancy. However, if the system can find another equilibrium state close to the original one, we say that there exists the neighborhood equilibrium. We evaluate the robustness of the equilibrium state by utilizing the rotating angle, where the system loses the neighborhood equilibrium. We show several numerical examples to verify our idea.
  • 福田 靖, 松日楽 信人, 戸田 義継, 町田 和雄, 岩田 敏彰
    1999 年 17 巻 1 号 p. 140-146
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
    This paper presents the locomotion of a space robot on a truss structure using its dual-arm manipulators. At the beginning of the locomotion on the truss structure, the space robot holds the truss with its dual-arm manipulators. And it releases one arm from the truss using the position control, and it grasps the truss using the target tracking and active limp control. Active limp control is introduced to release the force and torque between the manipulator and truss structure. It moves its body by the manipulator using the dual-arm control. And it releases the other arm from the truss and grasps it again. It continues above motion to walk on the truss structure. The effectiveness of the proposed method is shown in a simulation experiment of a locomotion on a truss structure on an air-bearing test-bed using the ground model of a space robot with a dual-arm manipulator.
  • 神野 誠, 栗田 勝裕
    1999 年 17 巻 1 号 p. 147-155
    発行日: 1999/01/15
    公開日: 2010/08/10
    ジャーナル フリー
    Force control of a manipulator is essential in a finishing robot system. But, there is not an effective force control method and a manipulator mechanism for every finishing task because finishing processes are different in individual tasks. In this paper, examination for deburring of machined tasted iron parts using a force controlled robot with a conic mill was performed. We have proposed a reference trajectory generate function (ROUND function). The ROUND function inserts an arc in the reference trajectory for the two straight edge parts. Thus the press direction changes continuously, and the robot can follow the edges. It is useful when a workpiece position and/or size is different from the taught one. Also, we have developed a new force control method and a tool hold mechanism for deburring of machined casted iron parts. In the force control method, a force feedback gain is proportional to force error. The tool hold mechanism make equalized compliance of the tool center point. The force feedback gain in the new method and the new tool hold mechanism was three times the gain in the conventional method and mechanism.
feedback
Top