检索结果-内蒙古大学图书馆

Intelligent learning control for a class of nonlinear dynamic systems

IEE PROCEEDINGS-CONTROL THEORY AND APPLICATIONS 1999年第2期146卷 165-170页

作者： Seo, WG Park, BH Lee, JS Pohang Univ Sci & Technol Dept Elect Engn Pohang 790784 South Korea Pohang Iron & Steel Co Tech Res Lab Instrumentat & Control Res Team Pohang South Korea

A controller is presented which guarantees system stability by using a feedback controller coupled with an intelligent compensator, and achieves precise tracking by using a set of iterative learning rules. In the feedback plus intelligent controller unit, the feedback control part stabilises the overall closed-loop system and keeps its error bounded, and the intelligent compensator estimates and compensates for the nonlinear part of the system, thereby keeping the feedback gains reasonably low in the feedback controller. In the iterative learning controller, a simple learning control rule is. used to achieve precise tracking of the reference signal and a parameter learning rule is used to update the parameters of the intelligent compensator, thereby identifying the uncertain nonlinearity as closely as possible.

关键词： stability intelligent control nonlinear dynamical systems learning systems learning control learning in ai Adaptive system theory closed-loop system Stability in control theory compensation closed loop systems tracking nonlinear dynamic systems Control system analysis and synthesis methods Nonlinear control systems feedback

来源：评论

学校读者我要写书评

暂无评论

Online unsupervised learning of hidden Markov models for adaptive speech recognition

引用

IEE PROCEEDINGS-VISION IMAGE AND SIGNAL PROCESSING 2001年第5期148卷 315-324页

作者： Chien, JT Natl Cheng Kung Univ Dept Comp Sci & Informat Engn Tainan 70101 Taiwan

A novel framework of an online unsupervised learning algorithm is presented to flexibly adapt the existing speaker-independent hidden Markov models (HMMs) to nonstationary environments induced by varying speakers, transmission channels, ambient noises, etc. The quasi-Bayes (QB) estimate is applied to incrementally obtain word sequence and adaptation parameters for adjusting HMMs when a block of unlabelled data is enrolled. The underlying statistics of a nonstationary environment can be successively traced according to the newest enrolment data. To improve the QB estimate, the adaptive initial hyperparameters are employed in the beginning session of online learning. These hyperparameters are estimated from a cluster of training speakers closest to the test environment. Additionally, a selection process is developed to select reliable parameters from a list of candidates for unsupervised teaming. A set of reliability assessment criteria is explored for selection. In a series of speaker adaptation experiments, the effectiveness of the proposed method is confirmed and it is found that using the adaptive initial hyperparameters in online learning and the multiple assessments in parameter selection can improve the recognition performance.

关键词： unsupervised learning adaptive signal processing learning in ai speech recognition speaker adaptation experiments Simulation, modelling and identification Bayes methods adaptive initial hyperparameters parameter estimation transmission channels test environment recognition performance Speech recognition unlabelled data Speech recognition and synthesis ambient noise reliability assessment criteria online unsupervised learning quasi-Bayes estimate speaker-independent HMM word sequence parameters online operation statistics hidden Markov models nonstationary environments enrolment data automatic speech recognition parameter selection Other topics in statistics adaptive speech recognition

来源：评论

学校读者我要写书评

暂无评论

Reactive parking control of nonholonomic vehicles via a fuzzy learning automaton

引用

IEE PROCEEDINGS-CONTROL THEORY AND APPLICATIONS 2001年第2期148卷 169-179页

作者： Rigatos, GG Tzafestas, SG Evangelidis, GJ Natl Tech Univ Athens Dept Elect & Comp Engn Intelligent Robot & Automat Lab GR-15773 Athens Greece

The paper presents an intelligent reactive controller, which is capable of performing automated parallel parking in a confined space slightly larger than the dimensions of the vehicle. This controller combines the features of fuzzy logic control and sliding-mode control and is modelled as a fuzzy learning, automaton. Starting with a brief exposition of the principles of reactive learning;and learning automata, the concept of the reduced complexity sliding-mode fuzzy logic controller (RC-SMFLC), employed, is introduced. Then the solution of the parking problem via the, RC-SMFLC, along with the convergence analysis, is presented. Representative simulation results are provided that demonstrate the parking steps and the effectiveness of the parking controller.

关键词： Automata theory convergence analysis learning in ai computational complexity sliding-mode control reactive parking control fuzzy control reactive learning Adaptive system theory convergence confined space reduced complexity sliding-mode fuzzy logic controller Computational complexity intelligent reactive controller nonholonomic vehicles automated parallel parking Neurocontrol variable structure systems Multivariable control systems learning automata fuzzy logic control fuzzy learning automaton Fuzzy control intelligent control RC-SMFLC Road-traffic system control automobiles

来源：评论

学校读者我要写书评

暂无评论

Iterative learning control: quantifying the effect of output noise

引用

IET CONTROL THEORY AND APPLICATIONS 2011年第2期5卷 379-388页

作者： Owens, D. H. Liu, S. Univ Sheffield Automat Control & Syst Engn Dept Sheffield S1 3JD S Yorkshire England

Fixed parameter iterative learning control (ILC) for linear-time invariant, single-input single-output systems subject to output noise is analysed with the intent of predicting the expectation of the underlying 'noise-free' mean square error (Euclidean norm) of the time series on each iteration. Explicit formulae are obtained in terms of the 'lifted' matrix models of the plant. Computational experiments are used to confirm the correctness of the proposed properties. Finally, frequency domain formulae are derived to provide insight into links between plant characteristics, noise spectra and other ILC parameters, and illustrated by application to the inverse-model-based ILC algorithm.

关键词： learning in ai time series Interpolation and function approximation (numerical analysis) frequency-domain analysis noise spectra output noise linear-time invariant Linear algebra (numerical analysis) matrix models single-input single-output systems iterative learning control mean square error methods learning (artificial intelligence) mean square error frequency domain formulae matrix algebra iterative methods

来源：评论

学校读者我要写书评

暂无评论

Hybrid learning approach to blind deconvolution of linear MIMO systems

引用

ELECTRONICS LETTERS 1999年第17期35卷 1429-1430页

作者： Choi, S Cichocki, A Chungbuk Natl Univ Dept Elect Engn Cheongju 361763 Chungbuk South Korea RIKEN Brain Sci Inst Lab Open Informat Syst Wako Saitama 35101 Japan

A hybrid network is presented which performs blind deconvolutions of linear MIMO systems. The hybrid network consists of a feedforward network followed by a feedback network, where each of the synapses is represented by an FIR filter. The FIR synapses in the feedforward network are learned by a Godard cost based algorithm and the FIR synapses in the Feedback network are updated by a spatio-temporal decorrelation algorithm so that different sources are recovered at different output nodes. An efficient spatio-temporal decorrelation algorithm based on the natural gradient is presented. The validity of the proposed method is confirmed by computer simulations.

关键词： Neural nets deconvolution FIR synapses hybrid network linear MIMO systems spatio-temporal decorrelation algorithm medical signal processing Signal processing theory filtering theory array signal processing digital communication hybrid learning approach natural gradient feedback feedforward neural nets feedforward network FIR filter FIR filters blind deconvolution learning (artificial intelligence) MIMO systems linear systems feedback network Neural computing techniques learning in ai Godard cost based algorithm Signal processing and detection decorrelation

来源：评论

学校读者我要写书评

暂无评论

learning object dynamics for smooth tracking of moving lip contours

引用

ELECTRONICS LETTERS 2000年第6期36卷 520-521页

作者： Wark, T Sridharan, S Chandran, V Queensland Univ Technol Sch Elect & Elect Engn RCSAVT Speech REs Lab Brisbane Qld 4001 Australia

A new technique is proposed for learning the dynamic characteristics of a deformable object applied in particular to the problem of lip-tracking. Experimental results are given which demonstrate that the use of dynamic models allows the system to track more robustly under adverse conditions and to correct spurious, poorly tracked frames.

关键词： adverse conditions learning (artificial intelligence) image motion analysis dynamics dynamic models dynamic characteristics learning moving lip contours learning in ai object dynamics learning tracking smooth tracking robust tracking Computer vision and image processing techniques deformable object

来源：评论

学校读者我要写书评

暂无评论

learning algorithm for pattern classification using cellular neural networks

引用

ELECTRONICS LETTERS 2000年第23期36卷 1941-1943页

作者： Grassi, G Di Sciascio, E Univ Lecce Dipartimento Ingn Innovazione I-73100 Lecce Italy Politecn Bari Dipartimento Elettrotecn & Elettron I-70125 Bari Italy

A new learning algorithm for pattern classification using cellular neural networks is described. The authors show that patterns belonging to the training set as well as patterns outside it can be classified reliably using the proposed algorithm. Comparisons with well established classification techniques clearly highlight the performances of the approach developed herein.

关键词： learning (artificial intelligence) Neural nets training set pattern classification cellular neural nets learning in ai classification techniques Pattern recognition learning algorithm cellular neural networks

来源：评论

学校读者我要写书评

暂无评论

Competitive neural network scheme for learning vector quantisation

引用

ELECTRONICS LETTERS 1999年第9期35卷 725-726页

作者： Wang, JH Peng, CY Natl Taiwan Ocean Univ Dept Elect Engn Keelung Taiwan

A novel self-development neural network scheme, which employs two resource counters to record node activity, is presented. The proposed network not only harmonises equi-error and equiprobable criteria, but it also avoids the stability-and-plasticity dilemma. Simulation results show that the new scheme displays superior performance (in terms of measured MSE, MAE, and training speed) over other neural network models.

关键词： equi-error criteria neural nets node activity Neural nets competitive neural network scheme vector quantisation unsupervised learning learning vector quantisation mean absolute error MAE learning in ai resource counters MSE equi-probable criteria self-development neural network scheme Interpolation and function approximation (numerical analysis) mean square error methods training speed

来源：评论

学校读者我要写书评

暂无评论

Optimal learning for patterns classification in RBF networks

引用

ELECTRONICS LETTERS 2002年第20期38卷 1188-1190页

作者： Hoang, TA Nguyen, DT Univ Tasmania Sch Engn Hobart Tas 7001 Australia

The proposed modifying of the structure of the radial basis function (RBF) network by introducing the weight matrix to the input layer (in contrast to the direct connection of the input to the hidden layer of a conventional RBF) so that the training space in the RBF network is adaptively separated by the resultant decision boundaries and class regions is reported. The training of this weight matrix is carried out as for a single-layer perceptron together with the clustering process. This way the network is capable of dealing with complicated problems, which have a high degree of interference in the training data, and achieves a higher classification rate over the current classifiers using RBF.

关键词： radial basis function networks learning (artificial intelligence) input layer clustering process Neural nets radial basis function network pattern classification class regions learning in ai training space Pattern recognition RBF networks classification rate improvement single-layer perceptron decision boundaries optimal learning weight matrix training

来源：评论

学校读者我要写书评

暂无评论

'Mechanical' neural learning for blind source separation

引用

ELECTRONICS LETTERS 1999年第22期35卷 1963-1964页

作者： Fiori, S Univ Perugia Dept Ind Engn I-06100 Perugia Italy Univ Ancona Dept Elect & Automat Ancona Italy

A class of learning models derived from the study of the dynamics of an abstract rigid mechanical system is presented. Application to blind source separation is illustrated through computer simulations.

关键词： learning (artificial intelligence) neural nets Neural nets computer simulations blind source separation signal sources signal detection learning in ai learning models neural learning Signal processing theory abstract rigid mechanical system dynamics

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：