检索结果-内蒙古大学图书馆

IEEE International Workshop on Intelligent Robots and Systems (IROS)

作者： J.C. Hudgens T. Arai Robotics Department Autonomous Machinery Division MITI/AIST Mechanical Engineering Laboratory Tsukuba Ibaraki Japan

A third generation parallel link manipulator (PLM) has been developed at the mechanical engineering laboratory. The device is similar in geometry to existing prototypes in the laboratory, however, the position sensing hardware configuration for online forward position analysis has been changed. The actuators and drivetrains are also more robust to provide for more realistic task demonstrations. This paper includes a brief description of the new prototype hardware. A closed-form forward position algorithm suitable for the new sensor configuration is derived, and a practical procedure for position sensor calibration is developed and demonstrated.

关键词： Prototypes Kinematics Manipulators Laboratories Hardware mechanical sensors mechanical engineering Geometry Actuators Robustness

来源：评论

学校读者我要写书评

暂无评论

Auto fuzzy tuning having minimum structure by using genetic algorithm and delta rule

Auto fuzzy tuning having minimum structure by using genetic ...

引用

International Conference on Emerging Technologies and Factory Automation (ETFA)

作者： T. Fukuda H. Ishigami T. Shibata F. Arai Department of Mechano-Informatics and Systems University of Nagoya Nagoya Japan Shin-Nagoya Thermal Power Station Cyubu Electric Power Company Inc. Nagoya Japan Bio-Robotics Division Robotics Department MITI Mechanical Engineering Laboratory Tsukuba Japan

An auto tuning algorithm of fuzzy inference for Fuzzy neural networks using the genetic algorithm and the delta rule is presented. Some auto-tuning methods are proposed to reduce time-consuming operations by human experts. This tuning method brings the minimal and optimal structure of the fuzzy model. Two types of the fuzzy model are prepared, whose membership functions on the antecedent part consist of triangular and Gaussian type, respectively. The effectiveness of the proposed methods compared with the former methods is shown by simulation. The proposed method has the potential to be applied to robotic motion control, sensing and recognition problems.< >

关键词： Genetic algorithms Humans Fuzzy systems Convergence Fuzzy neural networks Control systems Optimization methods Thermal engineering Power generation mechanical engineering

来源：评论

学校读者我要写书评

暂无评论

Dynamic memory by recurrent neural network and its learning by genetic algorithm

Dynamic memory by recurrent neural network and its learning ...

引用

IEEE Conference on Decision and Control

作者： T. Fukuda T. Kohno T. Shibata Department of Mechano-Informatics and Systems University of Nagoya Nagoya Japan Dept. of Mechano-Inform. & Syst. Nagoya Univ. Japan Bio-Robotics Division Robotics Department Mechanical Engineering Laboratory M.I.T.I. Tsukuba Japan

Recurrent neural networks have dynamic characteristics and can express functions of time. The recurrent neural networks can be applied to memorize robotic motions, i.e. trajectory of a manipulator. For this purpose, it is necessary to determine appropriate interconnection weights of the network. Formerly, learning algorithms based on gradient search techniques have been shown. However, it is difficult for the recurrent neural network to learn such functions while using previous approaches because of much computing requirement and limitation of memory. This paper presents a new learning scheme for the recurrent neural networks by genetic algorithm (GA). The GA is applied to determine interconnection weights of the recurrent neural networks. The GA approach is compared with the backpropagation through time which is a famous learning algorithm for the recurrent neural networks. Simulations illustrate the performance of the proposed approach.< >

关键词： Recurrent neural networks Genetic algorithms Neural networks Multi-layer neural network Neurons Feedforward neural networks Cost function Feedforward systems Robots mechanical engineering

来源：评论

学校读者我要写书评

暂无评论

Heuristic learning by genetic algorithm for recurrent neural network

Heuristic learning by genetic algorithm for recurrent neural...

引用

International Conference on Emerging Technologies and Factory Automation (ETFA)

作者： T. Fukuda T. Kohno T. Shibata Department of Mechano-Infomatics and Systems University of Nagoya Nagoya Aichi Japan Nagoya Daigaku Nagoya Aichi JP Dept. of Mechano-Inf. & Syst. Nagoya Univ. Japan Bio-Robotics Division Robotics Department Mechanical Engineering Laboratory Ministry of International Trade and Industry Tsukuba Ibaraki Japan

Recurrent neural networks have dynamic characteristics and express functions of time. Recurrent neural networks can memorize robotic motions, i.e., trajectories of manipulators For this purpose, it is necessary to determine appropriate interconnection weights of the network. A new learning scheme for the recurrent neural networks by genetic algorithm (GA) is presented. The GA is applied to determine interconnection weights of the recurrent neural networks. The proposed approach is compared with backpropagation through time for recurrent neural networks. Simulation illustrates the performance of the proposed approach.< >

关键词： Genetic algorithms Recurrent neural networks Neural networks Multi-layer neural network Neurons Cost function Electronic mail Feedforward neural networks Robot motion Manipulators

来源：评论

学校读者我要写书评

暂无评论

Finite element method in undergraduate engineering education

引用

Computer Applications in engineering Education 1993年第4期1卷 299-299页

作者： Nesbit, S.M. Comer, J.J. Van Gulick, L.A. Department of Mechanical Engineering Lafayette College Easton Pennsylvania 18042-1775 Dr. Steven M. Nesbit:is an assistant professor of Mechanical Engineering at Lafayette College where he has been teaching since 1990. He received his B.S. M.S. and Ph.D. degrees from West Virginia University. His areas of expertise are in mechanical design mechanisms and robotics. Previously he worked as a design engineer for Pratt-Whitney Aircraft. He is a member of ASME and ASEE. Dr. Jess Comer:is an assistant professor of Mechanical Engineering at Lafayette College. He received a B.S. and M.S. in Mechanical Engineering from South Dakota School of Mines and Technology and a Ph.D. from the University of Illinois at Urbana-Champaign. He was a co-author on the text “Fundamentals of Metal Fatigue Analysis” (Prentice-Hall 1990). Dr. Leonard A. Van Gulick:received his B.S. degree in Mechanical Engineering from Newark College of Engineering and his M.A. M.S. and Ph.D. degrees in Aerospace and Mechanical Sciences from Princeton University. He has been a staff engineer in the structural dynamics section at Martin Marietta Aerospace and a staff member in the Design Engineering Division at Los Alamos National Laboratory. Dr. Van Gulick is a professor of Mechanical Engineering at Lafayette College where he has also served as department head and program chair. Finite element analysis of medieval buildings is his primary research interest.

This article describes how the finite element method can be integrated into the education of undergraduate mechanical engineering students. The teaching of finite element theory, the use of commercial software in engineering courses, and the application of the finite element method in the context of mechanical design and analysis are discussed.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A predictive display with force feedback and its application to remote manipulation system with transmission time delay

A predictive display with force feedback and its application...

引用

1992 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 1992

作者： Kotoku, Tetsuo Cybernetics Division Robotics Department Mechanical Engineering Laboratory AIST-MITI Tsukuba Ibaraki305 Japan

ISBN: (纸本)0780307372

This paper discusses the design of a predictive display system with force feedback and its application to a remote manipulation system with transmission time delay. In the field of telerobotics, it has been found that an operator's performance is enhanced by the ability to obtain the information of remote environment with high fidelity. However, the transmission time delay produces one of the most crucial problems in space telerobotics. It is not practical to feed back force data continuously to the operator's hands because the delayed feedback may cause instability. In order to improve the manipulability, a system that presents the information of environments precisely to the operator without time delay is required. In this paper, a method for virtual generation of the constraining forces in the model world is proposed. As a first step, a simple robot and its environment model are constructed. It is assumed that the environment model consists of several polyhedral objects without friction on the surface and that the slave robot is a point object. The spring effects are assumed at the surface of the objects and then the constraining forces are produced in the direction normal to the surface, and those magnitudes are in proportion to the amount of the penetrations. A predictive display remote manipulation system with a force generation function is implemented. Some experiments were performed to confirm the effectiveness of the proposed method. © 1992 Institute of Electrical and Electronics Engineers Inc. All rights reserved.

关键词： Slave robots

来源：评论

学校读者我要写书评

暂无评论

A Predictive Display With Force Feedback And Its Application To Remote Manipulation System With Transmission Time Delay

A Predictive Display With Force Feedback And Its Application...

引用

IEEE International Workshop on Intelligent Robots and Systems (IROS)

作者： T. Kotoku Cybernetics Division Robotics Department Mechanical Engineering Laboratory AIST-MITI Tsukuba Ibaraki Japan

来源：评论

学校读者我要写书评

暂无评论

Estimation and control of the attitude of a dynamic mobile robot using internal sensors

引用

Advanced robotics 1992年第2期7卷 159-178页

作者： Matsumoto, Osamu Kajita, Shuuji Tani, Kazuo Mechanism Division Robotics Department Mechanical Engineering Laboratory MITI Ibaraki 305 Namiki 1-2 Tsukuba Japan

For the control of a dynamic mobile robot, the attitude in gravity space is an important state of the robot. Usually, the attitude is difficult to detect by simply using the signals from sensors. For example, an external sensor contacting the ground suffers disturbances from the roughness of the ground;the integration of a gyroscope signal has the problem of drift;an inclination sensor does not indicate the direction of gravity when acceleration exists. To solve these problems, we propose a control method in which the attitude of a mobile robot is estimated by an observer considering the robot dynamics and using only the information obtained by internal sensors. We applied this method to a wheeled inverted pendulum as an example of a dynamic mobile robot. The estimation of the attitude was made with good accuracy using the signals from the rate gyroscope and the motor encoder, and the control of stable running of the pendulum on a flat level plane worked successfully. We also realized the running control of a pendulum on an unknown rough road using the estimation of the slope gradient made by the observer. Thus, the effectiveness of the proposed method was demonstrated experimentally. © 1992 Taylor & Francis Group, LLC.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Force and Compliance Control

引用

Journal of robotics and Mechatronics 1991年第6期3卷 445-445页

作者： Tanie, Kazuo Bio-Robotics Division Robotics Department Mechanical Engineering Laboratory AIST-MITI 1-2 Namiki lbaraki Tsukuba305 Japan

来源：评论

学校读者我要写书评

暂无评论

High stiffness torque control for a geared DC motor based on acceleration controller

High stiffness torque control for a geared DC motor based on...

引用

International Conference on Industrial Electronics, Control and Instrumentation

作者： K. Kaneko N. Suzuki K. Ohnishi K. Tanie Cybernetics Division Robotics Department Mechanical Engineering Laboratory AIST MITI Tsukuba Ibaraki Japan Department of Mechanical Engineering Tamagawa University Machida Tokyo Japan Department of Electrical Engineering Keio University Yamato Japan Bio-Robotics Division Robotics Department Mechanical Engineering Laboratory AIST MITI Tsukuba Ibaraki Japan

The authors propose a torque control method for a geared DC motor with the ability to improve the structure stiffness reduced by the torque sensor. One of the special features of the proposed method is that it introduces an acceleration controller and positive feedback of the link angle through the inverse dynamics of the acceleration controller. The acceleration controller is realized by a disturbance observer which estimates the total sum of the disturbance torque imposed on a motor axis, and an acceleration controller is necessary for good performance of the positive feedback of the link angle. The effectiveness of the proposed method was demonstrated by experiments using a single axis geared DC motor system with a torque sensor.< >

关键词： Torque control DC motors Acceleration Servomotors Actuators Feedback mechanical engineering Robots Friction Laplace equations

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：