检索结果-内蒙古大学图书馆

IEEE Conference on Decision and control

作者： Georgios C. Chasparis Jeff S. Shamma Anders Rantzer Department of Automatic Control Lund University 221 00-SE Lund Sweden School of Electrical and Computer Engineering Georgia Institute of Technology Atlanta GA 30332

ISBN: (纸本)9781612848006

This paper presents a reinforcement learning algorithm and provides conditions for global convergence to Nash equilibria. For several reinforcement learning schemes, including the ones proposed here, excluding convergence to action profiles which are not Nash equilibria may not be trivial, unless the step-size sequence is appropriately tailored to the specifics of the game. In this paper, we sidestep these issues by introducing a new class of reinforcement learning schemes where the strategy of each agent is perturbed by a state-dependent perturbation function. Contrary to prior work on equilibrium selection in games, where perturbation functions are globally state dependent, the perturbation function here is assumed to be local, i.e., it only depends on the strategy of each agent. We provide conditions under which the strategies of the agents will converge to an arbitrarily small neighborhood of the set of Nash equilibria almost surely. We further specialize the results to a class of potential games.

关键词： Nash equilibrium learning automata perturbation function Agents Converge New classes global convergence Trivial Learning

来源：评论

学校读者我要写书评

暂无评论

Revealing Cluster Formation over Huge Volatile Robotic Data

Revealing Cluster Formation over Huge Volatile Robotic Data

引用

IEEE International Conference on Data Mining Workshops (ICDM Workshops)

作者： Nikos Mitsou Irene Ntoutsi Dirk Wollherr Costas Tzafestas Hans-Peter Kriegel Institute of Automatic Control Engineering Technische Universität München Germany School of Electrical and Computer Engineering Division of Signals Control and Robotics National and Technical University of Athens Greece Institute of Informatics Ludwig-Maximilians University of Munich Germany

In this paper, we propose a driven by the robotics field method for revealing global clusters over a fast, huge and volatile stream of robotic data. The stream comes from a mobile robot which autonomously navigates in an unknown environment perceiving it through its sensors. The sensor data arrives fast, is huge and evolves quickly over time as the robot explores the environment and observes new objects or new parts of already observed objects. To deal with the nature of data, we propose a grid-based algorithm that updates the grid structure and adjusts the so far built clusters online. Our method is capable of detecting object formations over time based on the partial observations of the robot at each time point. Experiments on real data verify the usefulness and efficiency of our method.

关键词： Vectors Three dimensional displays Robot sensing systems Robot kinematics Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

Neighborhood Discriminant Nearest Feature Line Analysis for Face Recognition

Neighborhood Discriminant Nearest Feature Line Analysis for ...

引用

International Conference on Innovations in Bio-inspired Computing and Applications (IBICA)

作者： Lijun Yan Jeng-Shyang Pan Shu-Chuan Chu John F. Roddick Department of Automatic Test and Control Harbin Institute of Technology Harbin China Department of Computer Science Shenzhen Graduate School Harbin Institute of Technology Shenzhen China School of Computer Science Engineering and Mathematics Flinders University of South Australia Australia

A novel subspace learning algorithm named neighborhood discriminant nearest feature line analysis (NDNFLA) is proposed in this paper. NDNFLA aims to find the discriminant feature of samples by maximizing the between-class feature line (FL) distances and minimizing the within-class FL distance. At the same time, the neighborhood is preserved in the feature space. Experimental results demonstrate the efficiency of the proposed algorithm.

关键词： Algorithm design and analysis Kernel Face Training Face recognition Principal component analysis Feature extraction

来源：评论

学校读者我要写书评

暂无评论

An improved KNN algorithm based on multi-attribute classification

引用

ICIC Express Letters, Part B: Applications 2011年第5期2卷 1117-1122页

作者： Zhang, Hongliang Li, Liangjun Li, Tienan Yang, Feng Computing Center of Anshan Normal University Anshan 114003 China School of Computer Science and Automatic Control Hebei United University No. 46 Xinhua Xidao Tangshan 063000 China School of Information Science and Engineering Northeastern University No. 11 Lane 3 Wenhua Road Heping District Shenyang 110819 China

To improve the classification accuracy of both the conventional Euclidean KNN algorithm and the improved KNN algorithm based on information entropy, this paper proposes an improved KNN algorithm based on multi-attribute classification. The procedures of the new algorithm comprise three steps: i) classify the attributes according to the percentage of their attribute values in an entire attribute of sample set into those discrete attributes suitable for entropy-based KNN algorithm and those continuous attributes suitable for conventional Euclidean KNN similarity-based algorithm;ii) process the two types of attributes separately and then sum up the two series of results with weighing and put the sum as the distance between samples;iii) select k samples which are closest to the test sample to determine the decision attribute type of the test sample. © 2011 ISSN 2185-2766.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Input Design for Nonlinear Stochastic Dynamic Systems – A Particle Filter Approach

引用

IFAC Proceedings Volumes 2011年第1期44卷 13191-13196页

作者： R. Bhushan Gopaluni Thomas B. Schön Adrian G. Wills Department of Chemical and Biological Engineering University of British Columbia Vancouver Canada V6T 1Z3 (Tel: 604 827 5668) Division of Automatic Control Linköping University SE-581 83 Linköping Sweden School of Electrical Engineering and Computer Science University of Newcastle Callaghan NSW 2308 Australia

Abstract We propose an algorithm for optimal input design in nonlinear stochastic dynamic systems. The approach relies on minimizing a function of the covariance of the parameter estimates of the system with respect to the input. The covariance matrix is approximated using a joint likelihood function of hidden states and measurements, and a combination of state filters and smoothers. The input is parametrized using an autoregressive model. The proposed approach is illustrated through a simulation example.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Tutorial: control issues in haptic teleoperation

Tutorial: Control issues in haptic teleoperation

引用

World Haptics Conference

作者： Angelika Peer Cristian Secchi Katsunari Sato Cenk Cavusoglu Institute of Automatic Control Engineering (LSR) Technische Universität München Munchen Germany DISMI University of Modena and Reggio Emilia and INFM Reggio Emilia Italy Graduate School of System Design and Management Keio University Yokohama Kanagawa Japan Department of Electrical Engineering and Computer Sciences Case Western Reserve University Cleveland OH USA

Telerobotics is one of the most traditional fields of robotics and it played a crucial role in the history of robotics and of the mankind, especially in the areas of space and undersea exploration and of remote material handling. On the other hand, teleoperation is still a very active research area and many problems are still open. In particular, the design of the control strategy for coupling local and remote site is of paramount importance for implementing telepresence, namely the feeling of being directly interacting with the remote environment. The IEEE RAS Technical Committee on Telerobotics would like to propose a half-day tutorial for illustrating several successful control strategies for implementing high performance bilateral teleoperation systems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Anti-cheating monitoring system of the electronic truck scale

Anti-cheating monitoring system of the electronic truck scal...

引用

2010 International Conference on Intelligent Computation Technology and Automation, ICICTA 2010

作者： Zhao, Yanjun Qu, Bin Gong, Ruikun Pan, Yifei College of Computer and Automatic Control Hebei Polytecnic University Tangshan China School of Electrical Engineering and Information Sichuan University Chengdu China

ISBN: (纸本)9780769540771

With the virtue of accuracy, fast and convenience, electronic truck scale is more and more applied to weighing measurement system of all walks of life. Since it is the very important measurement equipment in trade settlement, lawbreakers make many kinds of electronic truck scale cheating system to seek for illegal profit. Many cheating ways of electronic truck scale are analyzed in the paper. Probing deeply into the mechanism of current concealed cheating way-wireless remote control cheating method, a new anticheating method based on the wireless signal detection is put forward. If there is no wireless transmission signal detected, it can be judged that there is no cheating signal and continue to weigh. Otherwise, refuse to work and alarm. The wireless remote control signal detection system is designed. The specific frequency high-frequency modulated signals are detected in order to determine whether there have wireless cheating signals. The anti-cheating monitoring system of the electronics truck scale is designed. The experimental system is established and the experiment is carried out. The experimental results show that the method can real-time monitor the wireless cheating signals and the monitoring system can accurately judge whether there have cheating methods. © 2010 IEEE.

关键词： Monitoring

来源：评论

学校读者我要写书评

暂无评论

Spectral simulation analysis of doppler ultrasound blood flow signals from local expansion artery

Spectral simulation analysis of doppler ultrasound blood flo...

引用

International Forum on Information Technology and Applications

作者： Wang, Lifang Huang, Yaqun Tao, Yalin Zhou, Yaohua Department of Automatic Control and Mechanical Engineering Kunming University Kunming 650118 China Department of Computer Science Engineering Information School Yunnan University Kunming 650091 China School of Computer Science and Engineering Qujing Normal University Qujing 655011 China Department of Electronic Information and Control Engineering Guangxi University of Technology Liuzhou 545006 China

ISBN: (纸本)9780769541150

A simulation approach was proposed to analyze the spectra of Doppler signals in local expansion artery, which can provide a useful guidance for detecting the formation and growth progress of the aneurysms using the Doppler ultrasound technology and forecasting the size of aneurysm. Firstly, by solving Navier-Stokes equations, the velocity distributions in the vessels with various expansion degrees are calculated. Secondly, power spectral density (PSD) of the Doppler signals is estimated according to the relationship of velocity profiles and power spectral density. Finally, Doppler signals are generated using cosine-superposed methods. The results show that the spectra of Doppler signals are similar to those found in practice. Therefore, the proposed approach is useful for simulating the spectra of Doppler ultrasound signals. © 2010 IEEE.

关键词： Power spectral density

来源：评论

学校读者我要写书评

暂无评论

Study of the cost-sensitive AdaBoost face detection algorithm

Study of the cost-sensitive AdaBoost face detection algorith...

引用

2010 International Conference on computer Application and System Modeling, ICCASM 2010

作者： Song, Dingli Yang, Bingru Yu, Fuxing School of Information Engineering University of Science and Technology Beijing Beijing China College of Sciences Hebei Polytechnic University Tangshan Hebei Province China College of Computer and Automatic Control Hebei Polytechnic University China

ISBN: (纸本)9781424472369

It is the mainstream method that in human face detection and recognition with AdaBoost as the representative based on statistical learning method. Detection rates have reached a high level, and can achieve real-time detection. However, AdaBoost algorithm treats equally for different categories, there is no distinction between the cost of the different error categories. This paper presents a new cost-sensitive AdaBoost-based face detection algorithm, ensuring the detection rate and speed, effectively reducing the false detection rate, and improving the detection accuracy. © 2010 IEEE.

关键词： Adaptive boosting

来源：评论

学校读者我要写书评

暂无评论

Asymptotic suppression of a class of discrete-time hyperchaotic systems based on fuzzy control method

Asymptotic suppression of a class of discrete-time hyperchao...

引用

作者： Wu, Zhidan Pan, Ying Deng, Wei Song, Fusheng Fundamental Teaching Department of Computer and Mathematics Shenyang Normal University Shenyang 110034 China School of Mathematics and Computer Science Harbin University Harbin 150086 China Department of Automatic Control Engineering Shenyang Institute of Engineering Shenyang 110136 China

ISBN: (纸本)9787894631046

The chaotic suppression problem for a class of typical discrete-time hyperchaotic systems is investigated in this paper. A T-S fuzzy model-based control method is employed to eliminate hyperchaos in the considered discrete-time system. Firstly, the T-S fuzzy representation for the discrete-time hyperchaotic system is presented. Then, a state feedback controller is designed based on the parallel distributed compensation technique. According to the linear matrix inequality method and exact linearization technique, two kinds of approaches are given to solve the controller gain matrices. Finally, the validity of the proposed chaotic suppression method is illustrated by a numerical example.

关键词： Digital control systems

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：