检索结果-内蒙古大学图书馆

IEEE International Conference on Progress in Informatics and Computing (PIC)

作者： Wei Wang Xiaobo Lu Jiaji Song Cong Chen 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)9781509034857

This paper proposes a two-column convolutional neural network algorithm for face point detection, which proves better performance than single-column. Because of the deep level structure, global texture features are extracted at higher layers of the network, and the results of keypoint location remains high accuracy. In the first column, we take the R, G, B channel component of original image as input to train the neural network respectively and calculate the average location results. In the second one, we use Sobel operator to extract first-order derivative feature of original image and regard it as the input to train another convolutional network. The location results of two columns above are fused with appropriate proportion to ensure better results. Experimental result on LEW dataset shows the performance of the proposed parallel structure better than single convolutional network, and it stays robust due to occlusions, large pose variations and extreme lighting.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Human physical activity recognition based on computer vision with deep learning model

Human physical activity recognition based on computer vision...

引用

IEEE Instrumentation and measurement Technology Conference

作者： Lingfei Mo Fan Li Yanjia Zhu Anjie Huang School of Instrument Science and Engineering Southeast University Nanjing China Ministry of Education Key Laboratory of Measurement and Control of Complex Systems of Engineering Nanjing China

ISBN: (纸本)9781467392211

Human activity recognition is an active research area in the computer science because it is widely used in the fields of the security monitoring, health assessment, human machine interaction and other human related content searching. In this paper, a computer vision model based on the deep learning algorithm is proposed, which can recognize the human physical activity based on the skeleton data of the human body from the sensor of Microsoft Kinect. This model uses the human skeletons data from the CAD-60 dataset to recognize the human physical activity without using any prior knowledge. It can reduce the works on the stage of data preprocessing and feature extraction. It can also improve the generalization performance and robustness of the model, and give a better understanding of the human physical activity. Different tricks which can improve the performance of the neural networks, such as some regularization methods and other activation functions are tested. Finally, a convolutional neural network is used for the feature extraction, and a multilayer perceptron is used as the following classifier. The model can recognize twelve types of activities and the accuracy rate is 81.8%. It demonstrates that it is very effective to use the convolutional neural network to supervised learning and this model applies to human physical activity recognition.

关键词： Feature extraction Training Skeleton Neurons Computational modeling Biological neural networks Machine learning

来源：评论

学校读者我要写书评

暂无评论

Semi-Global Stabilization for a Class of Nonlinear Time-Delay systems by Linear Output Feedback

引用

IFAC-PapersOnLine 2015年第12期48卷 185-190页

作者： Wang, Pan Chai, Lin Fei, Shumin Liu, Guquan Key Laboratory of Measurement and Control of Complex Systems of Engineering Southeast University Ministry of Education School of Automation Nanjing 210096 China

The semi-global stabilization problem for a class of nonlinear systems with state time-delay is addressed in this paper. Sufficient conditions are derived by using Lyapunov-Krasovskii functional method and adjusting the scaling gain to guarantee the systems as well as the resulted closed-loop systems under a linear output feedback controller to be semi-globally stable. Numerical example is presented to illustrate the effectiveness of the obtained results in this paper. © 2015, IFAC (International Federation of Automatic control) Hosting by Elsevier Ltd. All rights reserved.

关键词： Closed loop systems

来源：评论

学校读者我要写书评

暂无评论

Spherical Formation Tracking control of Second-Order Nonlinear Agents With Directed Communication

Spherical Formation Tracking Control of Second-Order Nonline...

引用

IEEE International Conference on control and Automation

作者： Yang-Yang Chen Kai-Xuan Wang Ya Zhang School of Automation and Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University Chien-Shiung Wu College Southeast University

ISBN: (纸本)9781509017393

This article proposes a solution to the problem of forcing multiple agents to formation tracking a family of given concentric spheres when the communication topology is directed. Each agent under consideration has intrinsic second-order nonlinear dynamics. By using our previous concentric compression method, the cooperative control system can be regarded as a cascade system composed of the spherical tracking subsystem and the formation subsystem with the spherical tracking error as input. The algorithm for directed interaction is constructed by separately designing the control input projected onto the normal vector to the surface of sphere and the direction of spherical meridian for the spherical tracking subsystem, and the control input projected onto the direction of spherical parallel for formation motion. The asymptotic stability of the system is proved by using the tools of Input-to-State stability (ISS) theory. The effectiveness of the analytical results is verified by numerical simulations.

关键词： control input cascade systems Sphericity Normal vector SUB-SYSTEM Mentoring Agents tracking control System stability Nonlinear Dynamics PROBLEM SOLUTION

来源：评论

学校读者我要写书评

暂无评论

Global State control for a Class of Inherently Higher-order Parameterized Nonlinear Time-delay systems Based on Homogeneous Domination Approach

引用

IFAC-PapersOnLine 2015年第12期48卷 191-196页

作者： Chai, Lin Wang, Pan Fei, Shumin Key Laboratory of Measurement and Control of Complex Systems of Engineering Southeast University Ministry of Education School of Automation Nanjing 210096 China

This paper addresses the problem of global state feedback stabilization for a class of inherently higher-order parameterized nonlinear systems subject to time-delay. By using the homogeneous domination approach, we construct a homogeneous state feedback controller with an adaptive strategy. A constructive design method is developed based an effective coupling of the inductive method and a sign function, leading to adaptive regulators with minimal parameterization. With the aid of a homogeneous Lyapunov-Krasovskii functional, the closedloop systems can be asymptotically stabilized globally. © 2015, IFAC (International Federation of Automatic control) Hosting by Elsevier Ltd. All rights reserved.

关键词： Parameterization

来源：评论

学校读者我要写书评

暂无评论

Disturbance Observer Based Robust Finite control Set Predictive Torque control for Induction Motor systems

Disturbance Observer Based Robust Finite Control Set Predict...

引用

第35届中国控制会议

作者： Junxiao Wang Fengxiang Wang Shihua Li Ralph Kennel School of Automation Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Institute for Electrical Drive System and Power Electronics Technische Universitaet Muenchen

ISBN: (纸本)9781509009107

A disturbance observer(DOB) based finite control set predictive torque control(PTC) scheme for induction motors subject to lumped disturbances is proposed in this *** control an induction motor using the inverter switching character directly,the finite control set predictive torque control(PTC) method is studied which evaluates the electromagnetic torque and stator flux in the object *** optimized switching vector is choosed to control the insulated gate bipolar transistors(IGBTs),it also can minimize the error between reference signals and predicted *** to direct torque control(DTC),PTC method does not need a modulator and an internal current PI controller,but the speed loop also adopts the PI *** improving the disturbance reject ability and simplifying the PTC control structure,this work introduces a novel PTC method based on DOB technique which does not need the PI speed controller.

关键词： induction motor disturbance observer(DOB) predictive torque control(PTC) direct torque control(DTC) disturbance reject ability

来源：评论

学校读者我要写书评

暂无评论

Stability analysis of networked control systems with denial-of-service attacks

Stability analysis of networked control systems with denial-...

引用

UKACC International Conference on control (control)

作者： Jin Zhang Chen Peng Suhaib Masroor Hongtao Sun Lin Chai Shanghai University Shanghai Shanghai CN Shanghai Key Laboratory of Power Station Automation Technology Shanghai University Shanghai China Key Laboratory of Measurement and Control of Complex Systems of Engineering Southeast University Nanjing Jiangsu China

This paper addresses stability problem for a class of networked control systems (NCSs) with network attacks. As a network attack to control systems, denial-of-service attacks are fully taken into account in the NCSs modelling. Under the scenario that there exist denial-of-service attacks in the communication channel, a closed-loop system with an error function is provided for the studied NCSs, where the denial-of-service attacks are well considered. Then based on the above constructed system model, one sufficient condition is derived for ensuring the uniformly ultimately boundedness of the derived system under consideration, and the upper bound of the maximum allowable number of denial-of-service attack is obtained. Moreover, the celebrated time-delay system technology method is introduced to improve the result of the upper bound of the maximum allowable number of denial-of-service attack. Finally, one numerical example is given to show the effectiveness of the proposed method.

关键词： Computer crime Networked control systems Actuators Stability analysis Closed loop systems Communication channels

来源：评论

学校读者我要写书评

暂无评论

Adaptive Sliding Mode control for Re-entry Attitude of Near Space Hypersonic Vehicle Based on Backstepping Design

引用

IEEE/CAA Journal of Automatica Sinica 2015年第1期2卷 94-101页

作者： Jingmei Zhang Changyin Sun Ruimin Zhang Chengshan Qian the Flight Control Research Center Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education School of Automation Southeast University the College of Information and Control Nanjing University of Information Science & Technology

Combining sliding mode control method with radial basis function neural network (RBFNN), this paper proposes a robust adaptive control scheme based on backstepping design for re-entry attitude tracking control of near space hypersonic vehicle (NSHV) in the presence of parameter variations and external disturbances. In the attitude angle loop, a robust adaptive virtual control law is designed by using the adaptive method to estimate the unknown upper bound of the compound uncertainties. In the angular velocity loop, an adaptive sliding mode control law is designed to suppress the effect of parameter variations and external disturbances. The main benefit of the sliding mode control is robustness to parameter variations and external disturbances. To further improve the control performance, RBFNNs are introduced to approximate the compound uncertainties in the attitude angle loop and angular velocity loop, respectively. Based on Lyapunov stability theory, the tracking errors are shown to be asymptotically stable. Simulation results show that the proposed control system attains a satisfied control performance and is robust against parameter variations and external disturbances. © 2014 Chinese Association of Automation.

关键词： Airships Angular velocity Attitude control Backstepping control theory Design Functions Hypersonic aerodynamics Hypersonic vehicles Navigation Radial basis function networks Sliding mode control Uncertainty analysis Vehicles

来源：评论

学校读者我要写书评

暂无评论

Convergence Analysis of Consensus based Distributed Filtering Algorithm in Sensor Networks 34

Convergence Analysis of Consensus based Distributed Filterin...

引用

第三十四届中国控制会议

作者： ZHANG Ya TIAN Yu-Ping CAI Jun School of Automation Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education 73658 Army

ISBN: (纸本)9781467374439

This paper studies the distributed filtering problem of sensor networks, where a set of heterogeneous sensing nodes is required to estimate the state of a linear discrete-time dynamic system in a collaborative manner. Due to limited sensing range and communication constraints, just part of sensors can get observations of the process and a consensus based sub-optimal filtering algorithm is designed. The convergence properties of the estimation error covariance is further studied, and a necessary and sufficient convergence condition is provided based on LMIs. This condition shows that the convergence of the algorithm requires the target node in the extended topology of network is globally reachable and each sensing node is observable to the process. Simulation examples are given to illustrate the results.

关键词： Distributed filter,Sensor networks,Consensus,Convergence analysis,LMIs

来源：评论

学校读者我要写书评

暂无评论

Nonlinear Time-varying systems Identification Based on Multi-dimensional Taylor Network and Variable Forgetting Factor Recursive Least Squares Algorithm

引用

IFAC-PapersOnLine 2015年第11 11期48卷 1103-1107页

作者： Zhang, Jiao-Jun Yan, Hong-Sen School of Automation Southeast University Nanjing210096 China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing210096 China

A novel method for nonlinear time-varying systems identification based on multi-dimensional Taylor network and variable forgetting factor recursive least squares algorithm is proposed. In this paper, the connection weight coefficients of multi-dimensional Taylor network are regarded as time-varying parameters, which are trained by the variable forgetting factor recursive least squares algorithm, to reflect the input-output change of nonlinear time-varying systems. Simulation results show that the method proposed in this paper is valid. © 2015

关键词： Time varying networks Identification (control systems) Least squares approximations Time varying systems Connection weights Input output Multi dimensional Nonlinear time varying systems Recursive least squares algorithms Time varying parameter Variable forgetting factors

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：