检索结果-内蒙古大学图书馆

arXiv 2018年

作者： Wang, Gang Giannakis, Georgios B. Chen, Jie Digital Technology Center Department of Electrical and Computer Engineering University of Minnesota MinneapolisMN55455 United States State Key Lab of Intelligent Control Decision of Complex Systems School of Automation Beijing Institute of Technology Beijing100081 China Tongji University Shanghai200092 China

Neural networks with REctified Linear Unit (ReLU) activation functions (a.k.a. ReLU networks) have achieved great empirical success in various domains. Nonetheless, existing results for learning ReLU networks either pose assumptions on the underlying data distribution being e.g. Gaussian, or require the network size and/or training size to be sufficiently large. In this context, the problem of learning a two-layer ReLU network is approached in a binary classification setting, where the data are linearly separable and a hinge loss criterion is adopted. Leveraging the power of random noise perturbation, this paper presents a novel stochastic gradient descent (SGD) algorithm, which can provably train any single-hidden-layer ReLU network to attain global optimality, despite the presence of infinitely many bad local minima, maxima, and saddle points in general. This result is the first of its kind, requiring no assumptions on the data distribution, training/network size, or initialization. Convergence of the resultant iterative algorithm to a global minimum is analyzed by establishing both an upper bound and a lower bound on the number of non-zero updates to be performed. Moreover, generalization guarantees are developed for ReLU networks trained with the novel SGD leveraging classic compression bounds. These guarantees highlight a key difference (at least in the worst case) between reliably learning a ReLU network as well as a leaky ReLU network in terms of sample complexity. Numerical tests using both synthetic data and real images validate the effectiveness of the algorithm and the practical merits of the theory. Copyright © 2018, The Authors. All rights reserved.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

A Projection-Based Algorithm for Constrained L1- Minimization Optimization with Application to Sparse Signal Reconstruction

A Projection-Based Algorithm for Constrained L1- Minimizatio...

引用

International Conference on Information Science and Technology (ICIST)

作者： Qingshan Liu Wei Zhang Jiang Xiong Bingrong Xu Long Cheng School of Mathematics Southeast University Nanjing China College of Computer Science and Engineering Chongqing Three Gorges University Chongqing China School of Automation Huazhong University of Science and Technolog< Wuhan China State Key Lab. of Management and Control for Complex Systems University of Chinese Academy of Sciences Beijing China

In this paper, a projection-base algorithm is proposed for solving the constrained L 1 -minimization problem. Furthermore, the algorithm is utilized to sparse signal reconstruction. The L 1 -minimization is first converted into some equations which are described by the projections onto a hyper box set and the nonnegative quadrant. Then a iterative algorithm is proposed for solving the L 1 -minimization problem. Next, the algorithm is applied to sparse signal reconstruction described as an L 1 -minimization problem subject to L ∞ -norm noise constraint, or equivalently bound constraint. Finally, several experiments are presented to show the performance of the proposed algorithm.

关键词： Signal processing algorithms Signal reconstruction Optimization Convergence Electronic mail Recurrent neural networks Lyapunov methods

来源：评论

学校读者我要写书评

暂无评论

Design of controller system for industrial robot based on RTOS Xenomai 12

Design of controller system for industrial robot based on RT...

引用

12th IEEE Conference on Industrial Electronics and Applications, ICIEA 2017

作者： Yu, Chengjing Ma, Xudong Fang, Fang Qian, Kun Yao, Shun Zou, Yanping Key Laboratory of Measurement and Control of CSE School of Automation Southeast University Nanjing Jiangsu210096 China EandE Engineering Lab Center Southeast University Nanjing Jiangsu210096 China

ISBN: (纸本)9781538621035

A well-designed controller has a decisive effect on the performance of an industrial robot. Apart from control stability, good generality and flexible extension of the controller are becoming more and more important. In this paper, a multi-tasking robot controller software system based on Xenomai real-time operating system is presented. To ensure its generality, the controller system is multitasking designed into Robot control (RC), soft PLC, hardware driver, EtherCAT communication and management modules. The development for operation, synchronization, communication and cooperation among tasks and modules are given in detail, supporting a well working basic system. © 2017 IEEE.

关键词： controllers

来源：评论

学校读者我要写书评

暂无评论

Speed control of hydraulic elevator by using PID controller and self-tuning fuzzy PID controller

Speed control of hydraulic elevator by using PID controller ...

引用

Youth Academic Annual Conference of Chinese Association of automation (YAC)

作者： Xian Xu Qingling Wang Key Lab of Measurement and Control of Complex Systems of Engineering Southeast University Nanjing China

ISBN: (纸本)9781538629024

Hydraulic elevators are widely used in the world because of their simple structure, smooth running and comfortable ride, This paper presents a conventional PID controller and a self-tuning Fuzzy PID controller to control speed of elevator, their performance are compared by analyzing their step response curves, track-feature, error curves as well as ideal speed response curves. Simulation results show that Fuzzy PID controller has better control performance than the conventional PID controller. The modeling, control and simulation of the elevator system have been done using the software package MATlab/SIMULINK.

关键词： Hydraulic systems Elevators Niobium Automobiles control systems Fuzzy logic Transfer functions

来源：评论

学校读者我要写书评

暂无评论

Distributed Localization in Sensor Networks with Communication and measurement Noises

Distributed Localization in Sensor Networks with Communicati...

引用

Asian control Conference

作者： Fei Wu Yu-Ping Tian Bo Wang School of Automation Southeast University Nanjing China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing China

ISBN: (纸本)9781509015740;9781509015733

The localization problem for sensor networks with communication and measurement noises is studied in this paper. A robust distributed iterative algorithm called ECHO-CMN is presented based on the signed barycentric coordinate representation, which can be calculated only by relative distance measurements. In the algorithm, a dither is added to the sensor states before quantization to randomize the quantization error. To attenuate the communication noise, a gain parameter which decays to zero is used. An unbiased distance estimator is constructed, and nodes can take advantage of the estimator to update the distance estimation before each localization iteration. It can be proved that ECHO-CMN converges to the exact location of each sensor almost surely under some mild assumptions on the noise. Numerical studies illustrate the proposed localization algorithm.

关键词： Noise measurement Quantization (signal) Distance measurement Robustness Convergence Nickel Mathematical model

来源：评论

学校读者我要写书评

暂无评论

引用

Asian control Conference

作者： Siheng Zong Yu-Ping Tian School of Automation Southeast University Nanjing China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing China

ISBN: (纸本)9781509015740;9781509015733

This paper studies the similar formation algorithm of multi-agent systems with biased measurement errors. When relative position measurements contain biased errors, the existing similar formation algorithm is not mean reachable in general. This paper proposes a modified similar formation algorithm with which agents can overcome biased measurement errors online and reach any desired generic formation shape in the expectation sense. The formation error is globally uniformly bounded in the mean square sense.

关键词： measurement errors Position measurement Nickel Topology measurement uncertainty Algorithm design and analysis Eigenvalues and eigenfunctions

来源：评论

学校读者我要写书评

暂无评论

A New Processing Method of Infrared Temperature Images in Copper Electrolysis

A New Processing Method of Infrared Temperature Images in Co...

引用

Annual Conference of the IEEE Industrial Electronics Society

作者： Zhiying Hong Qingling Wang School of Automation Southeast University Nanjing China Ministry of Education Key Laboratory of Measurement and Control of Complex Systems of Engineering Nanjing China

In the copper electrolytic refining process, the short-circuit failure among the electrodes usually exists. When it occurs, the temperature distribution is often abnormal, accompanied by local or overall overheating. This paper proposes a novel processing method for infrared temperature images to effectively detect the short-circuit fault and locate the accurate malfunctioning spots. The design procedure contains contrast enhancement, filter denoising, edge detection, image segmentation, high temperature electrode positioning. Finally, experimental results show that the proposed method can guarantee real-time processing of the infrared image of the electrolytic cell and accurately mark the malfunctioning position of electrodes with high temperature fault.

关键词： Histograms Electrodes Copper Image edge detection Image enhancement Temperature distribution Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Estimation of image Jacobian Matrix with time-delay compensation based on Kalman filtering with colored noise 5

Estimation of image Jacobian Matrix with time-delay compensa...

引用

5th International Workshop on Advanced Computational Intelligence and Intelligent Informatics, IWACIII 2017

作者： Wang, Xinmei Liu, Zhenzhu Wei, Wu Liu, Feng School of Automation China University of Geosciences Hubei430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan Hubei430074 China College of Automation Science and Engineering South China University of Technology Guangzhou Guangdong510640 China

In order to compensate for the impact of time delay on the estimation of object feature point image and image Jacobian matrix and improve the control precision of image-based visual servo system, a Kalman filtering algorithm with colored noise is proposed to obtained the current motion state estimation of the feature point with time-delay compensation in the image space, then the accurate image Jacobian matrix under time delay can be obtained. For the above, the following work will be done in this paper: (1) the current motion state estimation of the feature point in the image space is predicted by Kalman filtering algorithm, combined with practical applications, measurement noise is usually as colored noise and can be regarded as a one-order Markov sequence excited by Gauss white noise, then a Kalman filtering recursive model can be derived according to the variance minimization principle of the estimation error;(2) to obtain the measurement vectors estimation which cannot be acquired in the Kalman filtering model owing to the existence of time delay, the relationship between the motion state of the feature point at the prior and later point is established, and a polynomial is fitted by using some of the observed data under the relationship to realize the measurement vectors estimation in the time-delay period. Thus the estimation of the motion state of the feature point at the current time in the image space can be obtained, and the estimation of image Jacobian matrix with time-delay compensation can be obtained. Experimental results validate the feasibility and superiority of this method.

关键词： Time delay

来源：评论

学校读者我要写书评

暂无评论

Adaptive Vi Be background model for vehicle detection

Adaptive Vi Be background model for vehicle detection

引用

2017 IEEE 第2届先进信息技术,电子与自动化控制国际会议(IAEAC 2017)

作者： Chengyi Pan Zhou Zhu Liangwei Jiang Min Wang Xiaobo Lu School of Automation Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education School of Automation Nanjing University of Science & Technology Key Laboratory of Ministry of Public Security for Road Traffic Safety

Background extraction is an important step in vehicle detection. In the actual scene, change of illumination will lead to a tremendous background change. It is necessary to update the background model

ISBN: (纸本)9781467389808

关键词： Background ViBe Adaptive Error function Evaluation conditions Vehicle detection Vehicle detection ERROR FUNCTIONS Self tuning BACKGROUND illumination Lighting

来源：评论

学校读者我要写书评

暂无评论

Coordinated Adaptive Flowfield Estimation for Second-Order Agents Formation Tracking Given Orbits 35

Coordinated Adaptive Flowfield Estimation for Second-Order A...

引用

第35届中国控制会议

作者： CHEN Yang-Yang TIAN Yu-Ping WANG Zan-Zan School of Automation Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University

ISBN: (纸本)9781509009107

This paper deals with cooperative estimation and control for second-order agents formation tracking a set of given orbit in an external flowfield,where only direction of flow inertial velocity is spatiotemporal and known.A novel coordinated adaptive estimator based on local neighbor-to-neighbor information is proposed to estimate the flow *** is shown that our previous geometric extension design,consensus and adaptive backstepping method can be combined together to construct the robust formation tracking controller under bidirectional *** theoretical result is proved by the numerical simulation.

关键词： Coordinated Adaptive Estimation Second-Order Agents Formation Tracking control

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：