检索结果-内蒙古大学图书馆

SICE Annual Conference

作者： Yuan-Pao Hsu Wei-Cheng Jiang Hsin-Yi Lin Department of Computer Science and Information Engineering National Formosa University Yunlin Taiwan Robotics Control Technology Department Intelligent Robotics Technology Division Industrial Technology and Research Institute Hsinchu Taiwan

In the this paper, a CMAC-Q-learning based Dyna agent is presented to relieve the problem of learning speed in reinforcement learning, in order to achieve the goals of shortening training process and increasing the learning, speed. We combine CMAC, Q-learning, and prioritized sweeping techniques to construct the Dyna agent in which a Q-learning is trained for policy learning; meanwhile, model approximators, called CMAC-model and CMAC-R-model, are in charge of approximating the environment model. The approximated model provides the Q-learning with virtual interaction experience to further update the policy within the time gap when there is no interplay between the agent and the real environment. The Dyna agent switches seamlessly between the real environment and the virtual environment model for the objective of policy learning. A simulation for controlling a differential-drive mobile robot has been conducted to demonstrate that the proposed method can preliminarily achieve the design goal.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Architecture of efficient RNS-based digital signal processor with very low-level pipelining

Architecture of efficient RNS-based digital signal processor...

引用

IET Irish Signals and Systems Conference (ISSC 2008)

作者： Stanislaw J. Piestrak Krzysztof S. Berezowski Universite de Lorraine - Site de Metz Metz Lorraine FR Institute of Computer Engineering Control and Robotics Wroclaw University of Technology Wroclaw Poland

ISBN: (纸本)9780863419317

A generalized architecture of an efficient digital signal processor using the residue number system (RNS) is proposed. It is based on using our new residue multipliers-accumulators (MACs) as the main building blocks. This architecture offers potentially higher throughput thanks to the possibility of implementing very low-level pipelining. The maximal applicable clock frequency could be determined by the delay of only a few stages of full-adders.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Design of residue multipliers-accumulators using periodicity

Design of residue multipliers-accumulators using periodicity

引用

IET Irish Signals and Systems Conference (ISSC 2008)

ISBN: (纸本)9780863419317

The design of multiplier-accumulators (MACs) that could be used to build digital filters using the residue number system (RNS) is considered. The generalized architectures of residue MACs modulo A built using carry-save adders (CSAs) are proposed. Unlike some earlier designs, the MACs are not limited to the moduli A of the form (2degplusmn1) and 2deg, which hence provides a designer with more choices of moduli A that could be used to form an RNS with the required dynamic range.

关键词：

来源：评论

学校读者我要写书评

暂无评论

CNC Machine Scheduling Problems with the Tool Aging

引用

IFAC Proceedings Volumes 2008年第3期41卷 122-126页

作者： Adam Janiak Wladyslaw Janiak Radoslaw Rudek Agnieszka Wielgus Institute of Computer Engineering Control and Robotics Wroclaw University of Technology Janiszewskiego 11/17 50-372 Wroclaw Poland Faculty of Computer Science and Management Wroclaw University of Technology Lukasiewicza 5 50-371 Wroclaw Poland

The aging effect can be understood as a deterioration of a machine (tool) that increases a time required to manufacture products. It is highly desirable to minimize the negative influence of this effect, which in case of CNC machine is usually compensated by the increasing of its working speed. Therefore, although a tool efficiency decreases the processing times of jobs (products) can be still kept on the desired level. On the other hand, the increasing of a working speed cases the faster decreasing of a durability of a tool and its premature failure. In this paper, we focus on optimization problems in manufacturing systems, where occur the tool wear of CNC drilling/cutting machine and its compensation at the cost of a durability of a tool. We model such problems by single machine scheduling problems with the aging effect and additional resources with two criteria: time criterion (makespan and maximum lateness) and resource consumption penalty. The objectives are the minimization of time criteria under a given resource consumption and the minimization of resource consumption under a given time criteria. Computational analysis of these problems together with solution algorithms are provided.

关键词： Scheduling Deteriorating Resource allocation CNC Tool wear

来源：评论

学校读者我要写书评

暂无评论

Accuracy Boosting Induction of Fuzzy Rules with Artificial Immune Systems

Accuracy Boosting Induction of Fuzzy Rules with Artificial I...

引用

International Multiconference on computer Science and Information Technology, IMCSIT

作者： Adam Kalina Edward Mezyk Olgierd Unold Value Based Advisors Sp. z o. o Wroclaw Poland Institute of Computer Engineering Control and Robotics Wroclaw University of Technology Wroclaw Poland

The paper introduces accuracy boosting extension to a novel induction of fuzzy rules from raw data using artificial immune system methods. Accuracy boosting relies on fuzzy partition learning. The modified algorithm w... 详细信息

关键词： Boosting Fuzzy systems Artificial immune systems Fuzzy sets Data mining Partitioning algorithms Genetic algorithms computer science Information technology control engineering computing

来源：评论

学校读者我要写书评

暂无评论

Single machine scheduling with job ready and delivery times subject to resource constraints

Single machine scheduling with job ready and delivery times ...

引用

International Symposium on Parallel and Distributed Processing (IPDPS)

作者： Wladyslaw Adam Janiak Adam Janiak Marie-Claude Portmann Faculty of Computer Science and Management Wrocław University of Technology Wroclaw Poland Institute of Computer Engineering Control and Robotics Wrocław University of Technology Wroclaw Poland Ecole des Mines de Nancy Nancy France

The paper deals with a single machine scheduling problem with job processing times dependent on continuously-divisible resource, e.g. gas, power, energy, raw materials, catalyzer, financial outlay. Ready and delivery times are also given for each job. The problem is to find a schedule of jobs and resource allocation that minimize the time by which all jobs are delivered. Genetic approach, using some proved problem properties is constructed to solve the problem being strongly NP-hard. Analysis of some computational experiment and conclusion remarks are also given.

关键词： Single machine scheduling Resource management Job shop scheduling Paper technology Raw materials Processor scheduling Genetics Manufacturing processes computer science Energy management

来源：评论

学校读者我要写书评

暂无评论

Investigation of fuzzy models for the valuation of residential premises using the KEEL tool

Investigation of fuzzy models for the valuation of residenti...

引用

8th International Conference on Hybrid Intelligent Systems, HIS 2008

作者： Lasota, Tadeusz Mazurkiewicz, Jacek Trawinski, Bogdan Trawinski, Krzysztof Agricultural University of Wroclaw Faculty of Environmental Engineering and Geodesy C.K. Norwida 25/27 50-375 Wroclaw Poland Wroclaw University of Technology Institute of Computer Engineering Control and Robotics Wybrzeze S. Wyspianskiego 27 50-370 Wroclaw Poland Wroclaw University of Technology Institute of Applied Informatics Wybrzeze S. Wyspianskiego 27 50-370 Wroclaw Poland Wroclaw University of Technology Faculty of Electronics Wybrzeze S. Wyspianskiego 27 50-370 Wroclaw Poland

ISBN: (纸本)9780769533261

A series of experiments aimed to generate and learn fuzzy models for the valuation of residential premises was conducted using the KEEL tool (Knowledge Extraction based on Evolutionary Learning). Four regression and four post-processing algorithms were applied to several data sets. They referred to sales/purchase transactions of residential premises, which were derived from the cadastral system and the registry of real estate transactions. The results proved the usefulness of the tool to carry out laborious and tedious investigation in a relatively short time. Further research is needed to determine to what extent data coming from such sources allow to build the real estate valuation models. © 2008 IEEE.

关键词： Housing

来源：评论

学校读者我要写书评

暂无评论

Multi-UAV Cooperative Fault Detection Employing Vision-Based Relative Position Estimation

引用

IFAC Proceedings Volumes 2008年第2期41卷 12093-12098页

作者： G. Heredia F. Caballero I. Maza L. Merino A. Viguria A. Ollero Robotics Computer Vision and Intelligent Control Group University of Seville Camino de los Descubrimientos s/n 41092 Seville Spain Pablo de Olavide University Crta. Utrera km. 1 41013 Seville Spain School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta 30332 USA

This paper presents a method to increase the reliability of UAV sensor fault detection in a multi-UAV context. The method uses additional position estimations that augment individual UAV fault detection system. These additional estimations are obtained using images from the same planar scene taken from two different UAVs. Since accuracy and noise level of the estimation depends on several factors, dynamic replanning of the multiUAV team can be used to obtain a better estimation in case of faults caused by slow growing errors of absolute position estimation that cannot be detected by using local FDI in the UAVs. Experimental results with data from two real UAVs are also presented.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Neural networks for disturbance and friction compensation in hard disk drives

Neural networks for disturbance and friction compensation in...

引用

IEEE Conference on Decision and control

作者： Chow Yin Lai Frank L. Lewis V. Venkataramanan Xuemei Ren Shuzhi Sam Ge Thomas Liew Centre for Life Sciences (CeLS) Graduate School for Integrative Sciences and Engineering (NGS) National University of Singapore Singapore Automation and Robotics Research Institute University of Texas Arlington Fort Worth TX USA Data Storage Institute Engineering Agency for Science Technology and Research Singapore Department of Automatic Control Beijing Institute of Technology Beijing China Department of Electrical and Computer Engineering National University of Singapore Singapore

ISBN: (纸本)9781424431236

In this paper, we show that the tracking performance of a hard disk drive actuator can be improved by using two adaptive neural networks, each of which is tailored for a specific task. The first neural network utilizes accelerometer signal to detect external vibrations, and compensates for its effect on hard disk drive position via feedforward action. In particular, no information on the plant, sensor and disturbance dynamics is needed in the design of this neural network disturbance compensator. The second neural network, designed to compensate for the pivot friction, uses a signum activation function to introduce nonlinearities inherent to pivot friction, thus reducing the neural network¿s burden of expectation. The stability of the proposed scheme is analyzed by the Lyapunov criterion. Simulation results show that the tracking performance of the hard disk drives can be improved significantly with the use of both neural networks compared to the case without compensation, or when only one of the networks is activated.

关键词： Neural networks Friction Hard disks Actuators Adaptive systems Feedforward neural networks Accelerometers Signal detection Stability analysis Stability criteria

来源：评论

学校读者我要写书评

暂无评论

Investigation of Fuzzy Models for the Valuation of Residential Premises Using the KEEL Tool

Investigation of Fuzzy Models for the Valuation of Residenti...

引用

International Conference on Hybrid Intelligent Systems (HIS)

作者： Tadeusz Lasota Jacek Mazurkiewicz Bogdan Trawinski Krzysztof Trawinski Institute of Applied Informatics Wroclaw University of Technology Wroclaw Poland Faculty of Environmental Engineering and Geodesy Agricultural University of Wrocław Wroclaw Poland Wroclaw University of Technology Institute of Computer Engineering Control and Robotics Wrocław Poland Wrocław University of Technology Institute of Applied Informatics Wrocław Poland Faculty of Environmental Engineering and Geodesy Agricultural University Wroclaw Poland Faculty of Electronics Wrocław University of Technology Wrocław Poland

A series of experiments aimed to generate and learn fuzzy models for the valuation of residential premises was conducted using the KEEL tool (knowledge extraction based on evolutionary learning). Four regression and four post-processing algorithms were applied to several data sets. They referred to sales/purchase transactions of residential premises, which were derived from the cadastral system and the registry of real estate transactions. The results proved the usefulness of the tool to carry out laborious and tedious investigation in a relatively short time. Further research is needed to determine to what extent data coming from such sources allow to build the real estate valuation models.

关键词： Data models Tuning Buildings Algorithm design and analysis Data mining Floors Cost accounting

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：