检索结果-内蒙古大学图书馆

11th International Conference on Learning Representations, ICLR 2023

作者： Yang, Sen Heng, Wen Liu, Gang Luo, Guozhong Yang, Wankou Yu, Gang School of Automation Southeast University China Key Lab of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University Nanjing China Tencent PCG China

In this paper we present a novel method to estimate 3D human pose and shape from monocular videos. This task requires directly recovering pixel-alignment 3D human pose and body shape from monocular images or videos, which is challenging due to its inherent ambiguity. To improve precision, existing methods highly rely on the initialized mean pose and shape as prior estimates and parameter regression with an iterative error feedback manner. In addition, video-based approaches model the overall change over the image-level features to temporally enhance the single-frame feature, but fail to capture the rotational motion at the joint level, and cannot guarantee local temporal consistency. To address these issues, we propose a novel Transformer-based model with a design of independent tokens. First, we introduce three types of tokens independent of the image feature: joint rotation tokens, shape token, and camera token. By progressively interacting with image features through Transformer layers, these tokens learn to encode the prior knowledge of human 3D joint rotations, body shape, and position information from large-scale data, and are updated to estimate SMPL parameters conditioned on a given image. Second, benefiting from the proposed token-based representation, we further use a temporal model to focus on capturing the rotational temporal information of each joint, which is empirically conducive to preventing large jitters in local parts. Despite being conceptually simple, the proposed method attains superior performances on the 3DPW and Human3.6M datasets. Using ResNet-50 and Transformer architectures, it obtains 42.0 mm error on the PA-MPJPE metric of the challenging 3DPW, outperforming state-of-the-art counterparts by a large margin. Code will be publicly available. © 2023 11th International Conference on Learning Representations, ICLR 2023. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Extended state observer based control for ground-coupled heat pump system 27

Extended state observer based control for ground-coupled hea...

引用

27th Chinese control and Decision Conference, CCDC 2015

作者： Yu, Zhang Wang, Junxiao Li, Shihua School of Automation Southeast University Nanjing Jiangsu Province China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education China

ISBN: (纸本)9781479970179

This paper aims to investigate an extended state observer (ESO) based control approach for the ground-coupled heat pump system (GCHP). As GCHP system has strong disturbances, a feasible approach of disturbance rejection should be proposed necessarily. ESO is employed to estimate the disturbance and compensate it to the controller to eliminate effects of the disturbance. By using ESO-based control and comparing with traditional control method, a better control performance is obtained through simulation verification. © 2015 IEEE.

关键词： disturbance rejection ESO GCHP simulation

来源：评论

学校读者我要写书评

暂无评论

Mean-square exponential stability for stochastic time-varying delay systems with Markovian jumping parameters:a delay decomposition approach

引用

Journal of systems engineering and Electronics 2011年第5期22卷 816-824页

作者： Li Ma Feipeng Da Key Laboratory of Measurement and Control for Complex Systems of Engineering Ministry of Education Research Institute of Automation Southeast University Nanjing 210096 E R. China College of Electronic and Information Engineering Jiangsu University Zhenjiang 212013 P. R. China

The mean-square exponential stability problem is investigated for a class of stochastic time-varying delay systems with Markovian jumping parameters. By decomposing the delay interval into multiple equidistant subintervals, a new delay-dependent and decay-rate-dependent criterion is presented based on constructing a novel Lyapunov functional and employing stochastic analysis technique. Besides, the decay rate has no conventional constraint and can be selected according to different practical conditions. Finally, two numerical examples are provided to show that the obtained result has less conservatism than some existing ones in the literature.

关键词： stochastic systems time-varying delay Markov chain linear matrix inequality （LMI）.

来源：评论

学校读者我要写书评

暂无评论

Weighted Group-Consensus of Multi-Agent systems with Bipartite Topologies 28

Weighted Group-Consensus of Multi-Agent Systems with Biparti...

引用

第28届中国控制与决策会议

作者： He Ying-gao Wang Xiang-yu Wuxi Institute of Arts & Technology School of Automation Southeast UniversityNanjing 210096China& Key Laboratory of Measurement and Control of Complex Systems of EngineeringMinistry of Education

ISBN: (纸本)9781467397155

This paper studies weighted group-consensus of multi-agent systems with bipartite topologies through adjusting the proportion of the current states and the delay states in the control algorithms.A sufficient condition for the convergence to a weighted group-consensus is proposed by adopting the generalized Nyquist stability criterion and the Gerschgorin disk theorem This condition is able to enlarge the upper bound on the maximum time-delay of weighted *** of the proposed result is demonstrated through numerical simulations.

关键词： Multi-agent systems group-consensus Nyquist criterion Gerschgorin disk theorem

来源：评论

学校读者我要写书评

暂无评论

Distributed solver for linear matrix inequalities: an optimization perspective

引用

control Theory and Technology 2021年第4期19卷 507-515页

作者： Weijian Li Wen Deng Xianlin Zeng Yiguang Hong Department of Automation University of Science and Technology of ChinaHefei 230027AnhuiChina Key Lab of Systems and Control Academy of Mathematics and Systems ScienceChinese Academy of SciencesBeijing 100190China Key Laboratory of Intelligent Control and Decision of Complex Systems School of AutomationBeijing Institute of TechnologyBeijing 100081China Department of Control Science and Engineering Tongji UniversityShanghai 201804China

In this paper,we develop a distributed solver for a group of strict(non-strict)linear matrix inequalities over a multi-agent network,where each agent only knows one inequality,and all agents co-operate to reach a consensus solution in the intersection of all the feasible *** formulation is transformed into a distributed optimization problem by introducing slack variables and consensus ***,by the primal–dual methods,a distributed algorithm is proposed with the help of projection operators and derivative ***,the convergence of the algorithm is analyzed,followed by illustrative simulations.

关键词： Distributed computation Distributed optimization Linear matrix inequalities Primal-dual method

来源：评论

学校读者我要写书评

暂无评论

Video Based Fall Detection Using Human Poses 9th

Video Based Fall Detection Using Human Poses

引用

9th CCF Conference on Big Data, BigData 2021

作者： Chen, Ziwei Wang, Yiye Yang, Wankou School of Automation Southeast University Nanjing210096 China Key Lab of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University Nanjing210096 China

ISBN: (纸本)9789811697081

Video based fall detection accuracy has been largely improved due to the recent progress on deep convolutional neural networks. However, there still exist some challenges, such as lighting variation, complex background, which degrade the accuracy and generalization ability of these approaches. Meanwhile, large computation cost limits the application of existing fall detection approaches. To alleviate these problems, a video based fall detection approach using human poses is proposed in this paper. First, a lightweight pose estimator extracts 2D poses from video sequences, and then 2D poses are lifted to 3D poses. Second, we introduce a robust fall detection network to recognize fall events using estimated 3D poses, which increases respective field and maintains low computation cost by dilated convolutions. The experimental results show that the proposed fall detection approach achieves a high accuracy of 99.83% on large benchmark action recognition dataset NTU RGB+D and real-time performance of 18 FPS on a non-GPU platform, 63 FPS on a GPU platform. © 2022, Springer Nature Singapore Pte Ltd.

关键词： Fall detection

来源：评论

学校读者我要写书评

暂无评论

Supervisory and energy management system of large public buildings

Supervisory and energy management system of large public bui...

引用

International Conference on Mechatronics and automation

作者： Ma, Xudong Cui, Ran Sun, Yu Peng, Changhai Wu, Zhishen Key Lab. of Measurement and Control of Complex Systems of Engineering School of Automation Southeast University Nanjing 210096 China International Institute for Urban Systems Engineering Southeast University Nanjing 210096 China

ISBN: (纸本)9781424451418

Buildings are important contributors to energy consumption accounting for around one-third of energy consumed in cities, where large public buildings are the dominant energy consumers and energy consumption might be significantly decreased through Building Energy Management systems (BEMS). A supervisory and management system for energy conservation in large public buildings is presented in this paper, with a PC or a embedded computer system running a BEMS software designed to connected to distributed energy metering system, wireless sensor networks, even conventional building automation (BA) systems. The supervisory and management software consists of information inquiry, energy consumption metering, energy efficiency analysis and decision support subsystems. The system framework is introduced with development and application of the software described in detail. © 2010 IEEE.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Image-Based Fast Finite-Time Target Tracking control of Onboard Inertially Stabilized Camera Platform

引用

IEEE Transactions on Industrial Electronics 2024年第12期71卷 16409-16418页

作者： Sun, Xunhong Zhu, Wenwu Du, Haibo Chen, Weile Chen, Chih-Chiang Hefei University of Technology School of Electrical Engineering and Automation Hefei230009 China Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing210096 China Southeast University School of Automation Nanjing210096 China National Cheng Kung University Department of Systems and Naval Mechatronic Engineering Tainan70101 Taiwan

In this article, we investigate the problem of fast finite-time target tracking control of the inertially stabilized platform (ISP) with a camera that mounted on a mobile robot. At first, the kinematics of the onboard ISP target tracking system is established by the geographic coordinate and pinhole camera model, which includes the state-dependent uncertainties and external disturbances. Then, based on the image information captured by camera, a continuous image-based fast finite-time control algorithm is designed for target tracking. Rigorous analysis proves that the target can be tracked at the center of camera's view in finite time in the absence of disturbances. In the presence of disturbances, the tracking error can converge to a region around the origin in finite time. Simulation and experimental comparisons are presented to confirm the efficiency of the proposed method. © 1982-2012 IEEE.

关键词： Mobile robots

来源：评论

学校读者我要写书评

暂无评论

Stochastic DoS Attack Allocation Against Collaborative Estimation in Sensor Networks

引用

IEEE/CAA Journal of Automatica Sinica 2020年第5期7卷 1225-1234页

作者： Ya Zhang Lishuang Du Frank L.Lewis School of Automation Southeast Universityand also with the Key Laboratory of Measurement and Control of Complex Systems of EngineeringMinistry of EducationNanjing 210096China University of Texas at Arlington Research Institute Fort WorthTX 76118 USA IEEE

In this paper,denial of service(DoS)attack management for destroying the collaborative estimation in sensor networks and minimizing attack energy from the attacker perspective is *** the communication channels between sensors and a remote estimator,the attacker chooses some channels to randomly jam DoS attacks to make their packets randomly dropped.A stochastic power allocation approach composed of three steps is ***,the minimum number of channels and the channel set to be attacked are ***,a necessary condition and a sufficient condition on the packet loss probabilities of the channels in the attack set are provided for general and special systems,***,by converting the original coupling nonlinear programming problem to a linear programming problem,a method of searching attack probabilities and power to minimize the attack energy is *** effectiveness of the proposed scheme is verified by simulation examples.

关键词： Attack allocation denial of service(DoS)attack packet loss remote estimation sensor networks

来源：评论

学校读者我要写书评

暂无评论

Apollonius Partitions Based Pursuit-evasion Game Strategies by Q-Learning Approach 41

Apollonius Partitions Based Pursuit-evasion Game Strategies ...

引用

第41届中国控制会议

作者： Qing Wang KaiQi Wu JianFeng Ye YongBao Wu Lei Xue SEU-Monash Joint Graduate school Southeast University The College of Software Engineering Southeast University School of Automation Southeast UniversityKey Laboratory of Measurement and Control of Complex Systems of EngineeringMinistry of Education

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

This paper studies a classical single pursuer and single evader pursuit-evasion *** pursuer attempts to capture the slower evader who aims to extend its lifetime during the *** simplify this question,requiring the evader to take fixation strategy which is choosing the farthest point in its current dominant region as aimpoint and moving at a constant *** the pursuer is faster than the ***,the speed ratio is a *** instaneous state space will be partitioned into pursuer's dominant zone and evader's dominant zone by the generalized Apollonius *** pursuit strategy is based on minimizing the area of the evader's dominant ***,we propose a supermodular game for this ***,the existence of the Nash equilibrium is *** results based on Q-learning are presented to solve the problem,which shows the effectiveness of this method.

关键词： Pursuit-evasion game Nash equilibrium Q-learning

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：