检索结果-内蒙古大学图书馆

Chinese control Conference

作者： Fei Long Leichao Pang Chuan-Ke Zhang Yong He Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan China

This paper investigates the stability of neural networks with a time-varying delay. Based on the good effectiveness of the augmented Lyapunov-Krasovskii functional (LKF), some useful integral vectors are summarized and used to construct single integral terms with augmented quadratic integrand so as to develop a novel augmented LKF candidate. Then an extended reciprocally convex matrix inequality and an auxiliary function-based inequality are utilized to estimate the derivative of the LKF. As a result, an improved stability criterion is established. Finally, the advantage of proposed method is demonstrated by a numerical example.

关键词： Linear matrix inequalities Stability criteria Symmetric matrices Biological neural networks Numerical stability Neurons

来源：评论

学校读者我要写书评

暂无评论

A Demand Analysis Method Based on TAKAGI-SUGENO Fuzzy Model for Drinking Service

A Demand Analysis Method Based on TAKAGI-SUGENO Fuzzy Model ...

引用

Chinese control Conference

作者： Man Hao Weihua Cao Zhentao Liu Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan China

A demand analysis method based on TAKAGI-SUGENO (T-S) fuzzy model for drinking service is proposed to provide corresponding services according to users' emotions and intentions in human-robot interaction, in which T-S fuzzy model is used to establish the relationship among human intention and human demands. First, the transformation of input and output is discussed. Secondly, fuzzy rules are formulated, and then fuzzy inference is applied to get user's demand corresponding to emotion and intention. The proposal considers peoples fuzziness in inferring humans intention, which could help the robots to provide satisfied drinking service to users. To validate the proposal, drinking service experiments are performed in a laboratory scenario using a humans-robots interaction system, from which the experimental results demonstrate the feasibility of the proposal.

关键词： Analytical models Human-robot interaction Takagi-Sugeno model Proposals Service robots

来源：评论

学校读者我要写书评

暂无评论

Observer-Based Tracking control for Suppressing Stick-Slip Vibration of Drillstring System

Observer-Based Tracking Control for Suppressing Stick-Slip V...

引用

Chinese control Conference (CCC)

作者： Jun Cheng Min Wu Luefeng Chen Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan China

A control method is proposed to suppress stick-slip vibration of drillstring system based on tracking control with zero steady-state error and state observer in this paper. A multiple degree-of-freedom (DOF) model of drillstring dynamic is presented first, which considers high-order modal of stick-slip vibration. Then, state observer is constructed to estimate the states of drillstring system, whose states are usually difficult to measure in the top of drillstring system. Finally, combing state feedback control and internal model principle, a tracking control with zero steady-state error is proposed to ensure the speed of rotary table and bit are consistent. The proposal only need top measurement, can eliminate multiple torsional modes, and has a strong robustness. Simulations show the effective of the proposal.

关键词： Vibrations Mathematical model Observers Torque Computational modeling State feedback

来源：评论

学校读者我要写书评

暂无评论

Dense Depth Estimation with Absolute Scale

Dense Depth Estimation with Absolute Scale

引用

Chinese control Conference (CCC)

作者： Xing Jin Zhiwen Yao Jingjing Zhang Hubei key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan China

Considering the difficulties in estimating depth from single image, in this paper, we propose a method to obtain the absolute scale depth map by combining the convolution neural network and depth filter. We compute relative transformation between consecutive frames by direct tracking features, which are extracted from RGB images and whose depthes are predicted by deep network, and then optimize relative motion by searching for a better feature alignment in epipolar line, and finally update every pixel depth of the reference frame by depth filter. We evaluate the proposed method on the open dataset comparison against the state of the art in depth estimation to evaluate our method.

关键词： Cameras Feature extraction Simultaneous localization and mapping Image reconstruction Pose estimation Neural networks

来源：评论

学校读者我要写书评

暂无评论

The transient electromagnetic inversion based on the simplex-simulated annealing algorithm

The transient electromagnetic inversion based on the simplex...

引用

Chinese control Conference (CCC)

作者： Guangjun Wang Caifeng Xu Gang Liu Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan China

In order to avoid the linear inversion method falling into local minima and slow convergence speed of the global optimization inversion method, the article proposed the simplex-simulated annealing algorithm for transient electromagnetic inversion research which combines advantages of the simplex method and the simulated annealing algorithm. The simplex method is used to obtain local minimum value which is relatively close to the actual value, then the simulated annealing algorithm is used to obtain the global optimal solution which can better reflect structural characteristics of the real stratigraphic model. Through the comparison of the noise inversion results and noise free inversion results about K-type, H-type, KH-type and HKH-type stratigraphic models, it can be proved that the simplex-simulated annealing algorithm can suppress some noise. The comparison of the simulated annealing method and the simplex-simulated annealing algorithm shows that the simplex-simulated annealing algorithm has the characteristics of global searching ability and fast convergence speed.

关键词： Simulated annealing Transient analysis Linear programming Geology Data models Convergence

来源：评论

学校读者我要写书评

暂无评论

Research on Handwritten Alphabet Recognition System Based on Extreme Learning Machine

Research on Handwritten Alphabet Recognition System Based on...

引用

Chinese control Conference

作者： Junlei Song Qi Liu Shengnan Tian Yi Wei Fang Jin Wenqin Mo Kaifeng Dong Hubei key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan China

Aiming at the problems of slow recognition, low efficiency and degree of automation in handwritten letter recognition system at present, a handwritten letter recognition system based on extreme learning machine is designed in this paper. The system is implemented by mixed programming with MATLAB and visual studio, it can reads, normalize, binarize and extract the handwritten letter images. The real-time interactive recognition of handwritten letters can be realized on the basis of training the simple pictures by using the identification model of the extreme learning machine algorithm. The experimental results show that the handwriting recognition system based on extreme learning machine designed in this paper can recognize 98.82% of handwritten letters and greatly reduce learning and testing time. Compared with BP neural network and other recognition algorithms, its training times have been reduced by hundreds or even thousands of times. At the same time, there is no manual intervention in the entire learning and testing process, which improves the automation of handwriting recognition.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Training and Testing Object Detectors With Virtual Images

引用

IEEE/CAA Journal of Automatica Sinica 2018年第2期5卷 539-546页

作者： Yonglin Tian Xuan Li Kunfeng Wang Fei-Yue Wang Department of Automation University of Science and Technology of China Hefei 230027 China State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijirtg 100190 China School of Automation Beijing Institute of Technology Beijing 100081. China Qingdao Academy of Intelligent Industries Qingdao 266000 China Research Center for Computational Experiments and Parallel Systems Technology National University of Defense Technology Changsha 410073 China IEEE

In the area of computer vision, deep learning has produced a variety of state-of-the-art models that rely on massive labeled data. However, collecting and annotating images from the real world is too demanding in terms of labor and money investments, and is usually inflexible to build datasets with specific characteristics, such as small area of objects and high occlusion level. Under the framework of Parallel Vision, this paper presents a purposeful way to design artificial scenes and automatically generate virtual images with precise annotations.A virtual dataset named Parallel Eye is built, which can be used for several computer vision tasks. Then, by training the DPM(Deformable parts model) and Faster R-CNN detectors, we prove that the performance of models can be significantly improved by combining Parallel Eye with publicly available real-world datasets during the training phase. In addition, we investigate the potential of testing the trained models from a specific aspect using intentionally designed virtual datasets, in order to discover the flaws of trained models. From the experimental results, we conclude that our virtual dataset is viable to train and test the object detectors.

关键词： Index Terms--Deep learning object detection parallel vision virtual dataset

来源：评论

学校读者我要写书评

暂无评论

Distributed circle formation control over directed networks with communication constraints ⁎

引用

IFAC-PapersOnLine 2019年第3期52卷 108-113页

作者： Peng Xu Jiayan Wen Chen Wang Guangming Xie College of Electrical and Information Engineering Guangxi University of Science and Technology Liuzhou 545006 China National Engineering Research Center for Software Engineering Peking University Beijing 100871 China The State Key Laboratory of Turbulence and Complex Systems Intelligent Biomimetic Design Lab College of Engineering Peking University Beijing100871 China

This paper investigates distributed circle formation problems of multi-agent systems (MAS) subject to limited information communication under a class of weight-unbalanced directed graphs, in which the communication topology contains a directed spanning tree and each agent can only perceive the distances from itself to the nearest neighbor in counterclockwise direction as well as the counterpart in the clockwise direction. Towards this end, a novel algorithm combined with an encoder-decoder framework has been proposed. We show that, under the proposed policy, the resulting network executions can drive the states of all mobile agents to converge to some expected equilibrium point. Numerical simulation results have been given to demonstrate the effectiveness of the proposed algorithm.

关键词： Multi-agent systems Communication Bandwidth Directed Network Circle Formation control

来源：评论

学校读者我要写书评

暂无评论

A Two-level Adaptive Target Recognition and Tracking Method Based on Vision for Multi-robot System

A Two-level Adaptive Target Recognition and Tracking Method ...

引用

IEEE International Conference on Robotics and Biomimetics

作者： Liang Ren Zhiqiang Cao Min Tan Peng Zhao Xuechao Chen State Key Laboratory of Management and Control for Complex Systems University of Chinese Academy of Sciences Beijing China The 21st Research Institute of China Electronics Technology Group Corporation Shanghai China Beijing Advanced Innovation Center for Intelligent Robots and Systems BIT Beijing China

The vision-based target recognition and tracking have received much attention in the field of robotics. Existing methods mainly focus on the vision perception of individual robot with a single view, however, the performance is susceptible to illumination and occlusion. Multi-robot collaborative perception provides a potential solution to deal with the limitation of single-view observation, however, the challenging of environmental adaptability for multi-robot collaborative decision still remains unsolved. To solve this problem, this paper proposes a two-level adaptive target recognition and tracking method based on vision for multi-robot system. The problem of multi-robot target recognition and tracking is solved under a two-level framework, which contains the features fusion level of individual robot and the cooperation level of multi-robot system. In the first level, the features measuring results that influence the visual perception of individual robot are fused, while the second level combines the voting of each robot together to determine the target for multi-robot system. Both the features measuring weights and robots voting weights are adaptively updated according to their evaluation, which lead to a beneficial result where the features and robots with higher accuracy play major roles in the first and second levels, respectively. Therefore, a good adaptability to the environments can be guaranteed. The experimental results show that the proposed approach can realize the coordination of multi-robot system in target recognition and tracking with an effective performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Adaboost-kNN for dynamic emotion recognition 8

Adaboost-kNN for dynamic emotion recognition

引用

8th International Symposium on Computational Intelligence and Industrial Applications and 12th China-Japan International Workshop on Information Technology and control Applications, ISCIIA and ITCA 2018

作者： Li, Min Su, Wanjuan Chen, Luefeng Wu, Min Hirota, Kaoru School of Automation China University of Geosciences Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Tokyo Institute of Technology Yokohama226-8502 Japan

— K-nearest Neighbor based adaptive boosting (AdaBoost-KNN) is proposed for emotion understanding in human-robot interaction (HRI), where the real-time dynamic emotion is recognized according to facial expression. It can make robots capable of understanding human emotions and make HRI run smoothly. The Candide-3 model was adopted to extract the two-dimensional coordinate values of the key points of facial emotion, as the basis for emotion classification. The emotion classification is based on AdaBoost-KNN, which sets up seven basic KNN classifier. Then it iterates seven times to update the weight of samples, and combines the weak classifier with coefficient to establish the final classifier. Finally, the performance of the proposal is verified by using K-fold cross validation. Results show that the proposal achieves higher accuracy than that of a single KNN. © ISCIIA and ITCA 2018 - 8th International Symposium on Computational Intelligence and Industrial Applications and 12th China-Japan International Workshop on Information Technology and control Application.

关键词： Adaptive boosting

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：