检索结果-内蒙古大学图书馆

13th ieee data driven control and learning systems conference, DDCLS 2024

作者： Wang, Li Qu, Qilin Wang, Yanwei Wong, David Shan-Hill Zheng, Ying School of Artificial Intelligence and Automation Huazhong University of Science and Technology China -Belt and Road Joint Laboratory on Measurement and Control Technology Wuhan430074 China School of Mechanical & Electrical Engineering Wuhan Institute of Technology 693 Chuxiong Road Wuhan430073 China National Tsing-Hua University Department of Chemical Engineering Hsinchu30071 Taiwan

ISBN: (纸本)9798350361674

Imbalance in fault samples is a key problem with most rolling bearing datasets in real industrial processes, limiting the performance of fault detection. In addition, existing fault diagnosis methods for rolling bearings do not adopt a modular division between the initial layer and the classification layer. This lack of modularization inhibits the full activation of the network's feature extraction capabilities and impedes subsequent model improvements. In this paper, a novel fault diagnosis method based on data augmentation integrated with a feature-enhanced convolutional neural network (DA-FECNN) is proposed. First, Variational Autoencoder-Generative Adversarial Network (VAE-GAN) is utilized to expand the number of fault samples and improve the diversity of few-shot. Second, a uniform feature-enhanced structure is designed with a feature extraction module, a multi-scale classification module, and a reconstruction module. The network's ability on feature extraction is improved by focusing on high quality features. Thereby, accurate fault classification is realized simultaneously. Experiments on real world data show the effectiveness of the proposed method on fault diagnosis task. © 2024 ieee.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A Cascaded Broad learning System for Manipulator Motion control

A Cascaded Broad Learning System for Manipulator Motion Cont...

引用

2024 International conference on Intelligent Robots and systems

作者： Zuo, Guoyu Dong, Shuaifeng Zhou, Jiyong Yu, Shuangyue Beijing Univ Technol Fac Informat Technol Beijing 100124 Peoples R China Beijing Key Lab Comp Intelligence & Intelligent S Beijing 100124 Peoples R China

ISBN: (纸本)9798350377712;9798350377705

Intelligent control methods have led to a significant simplification of the robotic arm modeling and control tuning process, and thus they have been widely used. To further improve the precision of robotic arm motion control, this paper proposes a robotic arm motion control strategy based on a cascaded feature-enhanced elastic-net broad learning system (CFE-EN-BLS). This will fully extract data features to improve motion control accuracy. Moreover, ElasticNet regression is introduced to reduce feature redundancy. Finally, Lyapunov stability theory is introduced to constrain the learning parameters of the proposed learning method to enhance the convergence of the control strategy. The simulation and experiment show that the proposed control strategy can realize high-precision trajectory tracking control of the robotic arm.

关键词： Robotic arms

来源：评论

学校读者我要写书评

暂无评论

Reinforcement learning for Non-Deterministic Transition systems With an Application to Symbolic control

引用

ieee control systems LETTERS 2023年 7卷 1610-1615页

作者： Borri, Alessandro Possieri, Corrado CNR Ist Anal Sistemi & Informat A Ruberti I-00185 Rome Italy Univ Aquila Ctr Excellence Res DEWS I-67100 Laquila Italy Univ Roma Tor Vergata Dipartimento Ingn Civile & Ingn Informat I-00133 Rome Italy

Reinforcement learning (RL) is a well-established framework for the computation of optimal control policies maximizing the expected reward collected along the evolution of Markov decision processes. In this letter, we extend the RL framework to non-deterministic finite transition systems (FTSs), whose solutions are non-unique but not endowed with a probability measure. We show how to dynamically build RL controllers (possibly learning the FTS model just from experience) maximizing the best-case and worst-case return obtained from a trajectory (run) of the model, assuming full-state information. The framework is successfully applied to the case in which the considered transition system is obtained as a finite approximation of a continuous system, also called a symbolic model. Numerical results on the classical mountain car benchmark highlight the potential of the proposed approach.

关键词： Mathematical models Reinforcement learning Heuristic algorithms Optimal control Numerical models Computational modeling Buildings Automata optimal control data driven control reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

data-driven Optimal Formation-Containment control for a Group of Spacecrafts Subject to Switching Topologies

引用

ieee TRANSACTIONS ON AEROSPACE AND ELECTRONIC systems 2023年第5期59卷 6087-6097页

作者： Cheng, Ming Liu, Hao Wan, Yan Valavanis, Kimon P. Lewis, Frank L. Beihang Univ Sch Astronaut Beijing 100191 Peoples R China Beihang Univ Inst Artificial Intelligence Beijing 100191 Peoples R China Zhongguancun Lab Beijing Peoples R China Univ Texas Arlington Dept Elect Engn Arlington TX 76010 USA Univ Denver Dept Elect & Computer Engn Denver CO 80208 USA Univ Texas Arlington Univ Texas Arlington Res Inst Ft Worth TX 76118 USA

The research focuses on the data-driven optimal time-varying formation-containment control problem for a group of heterogeneous spacecrafts that are simultaneously subjected to switching topologies, nonlinear relative motion dynamics, and parameter variations. A distributed observer is developed for each spacecraft to generate the position references and to decrease the influences of switching topologies. The optimal formation-containment control protocol is learned using relative motion data and system inputs, based on the reinforcement learning theory. Simulations of a spacecraft group are given to show the performance of the proposed formation-containment control protocol.

关键词： Space vehicles Topology Switches Protocols Aerodynamics Optimal control Observers Aerospace control formation-containment control intelligent control optimal control nonlinear system

来源：评论

学校读者我要写书评

暂无评论

data driven Decentralized control of Inverter Based Renewable Energy Sources Using Safe Guaranteed Multi-Agent Deep Reinforcement learning

引用

ieee TRANSACTIONS ON SUSTAINABLE ENERGY 2024年第2期15卷 1288-1299页

作者： Zhang, Mengfan Guo, Guodong Magnusson, Sindri Pilawa-Podgurski, Robert C. N. Xu, Qianwen KTH Royal Inst Technol Elect Power & Energy Syst Div S-11428 Stockholm Sweden Stockholm Univ Dept Comp & Syst Sci S-11419 Stockholm Sweden Univ Calif Berkeley Dept Elect Engn & Comp Sci Berkeley CA 94720 USA

The wide integration of inverter based renewable energy sources (RESs) in modern grids may cause severe voltage violation issues due to high stochastic fluctuations of RESs. Existing centralized approaches can achieve optimal results for voltage regulation, but they have high communication burdens;existing decentralized methods only require local information, but they cannot achieve optimal results. Deep reinforcement learning (DRL) based methods are effective to deal with uncertainties, but it is difficult to guarantee secure constraints in existing DRL training. To address the above challenges, this paper proposes a projection embedded multi-agent DRL algorithm to achieve decentralized optimal control of distribution grids with guaranteed 100% safety. The safety of the DRL training is guaranteed via an embedded safe policy projection, which could smoothly and effectively restrict the DRL agent action space, and avoid any violation of physical constraints in distribution grid operations. The multi-agent implementation of the proposed algorithm enables the optimal solution achieved in a decentralized manner that does not require real-time communication for practical deployment. The proposed method is tested in modified ieee 33-bus distribution and compared with existing methods;the results validate the effectiveness of the proposed method in achieving decentralized optimal control with guaranteed 100% safety and without the requirement of real-time communications.

关键词： Voltage control Safety Renewable energy sources Uncertainty Reinforcement learning Real-time systems Optimization Inverter based renewable energy sources deep neural network deep reinforcement learning safe decentralized control

来源：评论

学校读者我要写书评

暂无评论

Deep Metric learning with Impulsive Neural Network 13

Deep Metric Learning with Impulsive Neural Network

引用

13th ieee data driven control and learning systems conference, DDCLS 2024

作者： Chen, Hong-Zhou Hu, Bin Chen, Long Guan, Zhi-Hong School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China School of Future Technology South China University of Technology Guangzhou510641 China School of Electrical and Information Engineering Wuhan Institute of Technology Wuhan430205 China

ISBN: (纸本)9798350361674

Neural networks with siamese and triplet structures can discriminate between samples by explicitly learning the distance measure. But the dominant metric learning model based on deep neural networks (DNN) has issues with low energy efficiency and slow on-chip operating speed. Motivated by this, we put forth novel impulsive neural networks (INN) based deep metric learning approachs. The loss function is proposed based on the Euclidean distance between the membrane potentials of the neurons in the output layer of the sub-networks composed by Leaky integrate-and-fire (LIF) neurons. Furthermore, the back-propagation through time (BPTT) technique can be effectively implemented and utilized for network optimization through the surrogate functions. Experiments on multiple datasets show that the proposed model can learn the difference between images with performance very close to DNN-based models. At the same time, we conducted an intuitive energy consumption analysis of our model. The results show that compared with DNN-based models, the proposed model has higher energy efficiency. Besides, it only needs an extremely short time step T=10 to complete inference process, which makes the low-latency on-chip operation possible. © 2024 ieee.

关键词： Energy Efficiency Image Processing Impulsive Neural Network Metric learning

来源：评论

学校读者我要写书评

暂无评论

Application of Bilateral control Based on Additive Decomposition to Image Rotation Compensation 13

Application of Bilateral Control Based on Additive Decomposi...

引用

13th ieee data driven control and learning systems conference, DDCLS 2024

作者： Sun, Xinchao Ran, Yan Wang, Jiachen Zhao, Liyun Qin, Wei Li, Xinwei School of Automation and Electrical Engineering Inner Mongolia University of Science and Technology China Key Laboratory of Integrated Automation for Process Industries Universities of Inner Mongolia Autonomous Region Baotou014000 China

ISBN: (纸本)9798350361674

This paper focuses on the image rotation compensation servo mechanism in aerial imaging systems and the influence of disturbances on its tracking accuracy. Meanwhile, a bilateral control structure based on servo motor drive is utilized to realize the image rotation compensation, and the additive decomposition theory is employed to decompose the system into two subsystems, where the main system is responsible for accurate tracking, and the auxiliary system is designed with a generalized nonlinear extended state observer (G-NESO) to observe and compensate for equivalent perturbations, which is combined with a novel higher-order sliding mode compensator to refine further and compensate the system perturbations. The stable tracking of a desired trajectory and the consistent output of the two systems are achieved, and the image rotation problem caused by the motion asynchrony between the scanning mirror and the imaging plane is solved. The stability of the control system is demonstrated using Lyapunov theory. Finally, MATLAB simulation is utilized to verify the effectiveness and feasibility of the method. © 2024 ieee.

关键词： State estimation

来源：评论

学校读者我要写书评

暂无评论

On-Orbit Identification of Unstable Micro-Propulsion System Parameters Using Drag-Free control 13

On-Orbit Identification of Unstable Micro-Propulsion System ...

引用

13th ieee data driven control and learning systems conference, DDCLS 2024

作者： Xu, Nuo Xia, Xiwang Li, Zhaoxiong He, Xiongfeng Guo, Ning Wang, Pengcheng Zhang, Yonghe Chinese Academy of Sciences Innovation Academy for Microsatellites Shanghai201203 China University of Chinese Academy of Sciences Beijing100049 China Key Laboratory for Satellite Digitization Technology Chinese Academy of Sciences Shanghai201203 China Lanzhou Institute of Physics Science and Technology on Vacuum Technology and Physics Lab Lanzhou730000 China

ISBN: (纸本)9798350361674

Space gravitational wave detection spacecraft is equipped with a drag-free control and micro-propulsion system, and adopts thrust closed-loop feedback control structure. An on-orbit high accuracy estimation algorithm of micro-thruster parameters based on drag-free control model is proposed in this paper in order to deal with the problem of unsteady nominal thrust caused by thruster failure or aging and and extend thrusters lifespan. In this method, the recursive gain and covariance are designed to estimate the parameter matrix of system. Adaptive forgetting factors are added to the covariance matrix for designing a fading memory window to determine the influence of prior data on the results. At the same time, the thruster state is deconstructed into three parts: thrust gain, thrust direction and lever arm to facilitate the redistribution of control force and torque. Simulations show that the on-orbit thrust identification errors of the proposed method in different micro-propulsion states are all at sub-micro-newton level, and the noise is less than 0.1μ N/Hz which has capability to meet the requirements corresponding to drag-free control system. © 2024 ieee.

关键词： Covariance matrix

来源：评论

学校读者我要写书评

暂无评论

EMLFCL: An Efficient Multilevel Fusion Contrastive learning for Multiview Clustering

引用

ieee TRANSACTIONS ON NEURAL NETWORKS AND learning systems 2025年 PP卷 PP页

作者： Yi, Yugen Zhang, Ningyi Zhang, Zehui Fu, Yijian Chen, Lei Wang, Jianzhong Jiangxi Normal Univ Sch Software Nanchang 330022 Jiangxi Peoples R China Jiangxi Prov Engn Res Ctr Blockchain Data Secur & Nanchang 330022 Jiangxi Peoples R China BoulderAI Technol Co Ltd Hangzhou 311215 Zhejiang Peoples R China Northeast Normal Univ Coll Informat Sci & Technol Changchun 130117 Jilin Peoples R China

Multiview clustering (MVC) with contrastive learning (CL) has attracted considerable interest. Nevertheless, current methods have specific drawbacks since the coherence between views in them is limited either at the feature representation level or the cluster representation level. Besides, certain methods demonstrate subpar performance and limited robustness when handling noisy data. This article introduces an efficient multilevel fusion CL framework for MVC called EMLFCL. The EMLFCL model seamlessly incorporates a shared multi-layer perceptron (MLP) network (MNet) and a fusion network (FNet) to capture and merge common representation information, which effectively eliminates the impact of view-specific private information during the clustering process. Specifically, we establish an efficient multilevel CL strategy at both the feature representation level and the clustering representation level. Rather than rely on pairwise comparisons between views, our proposed CL strategy makes comparisons between different views and the anchor view. Since the anchor view contains abundant shared information, this strategy effectively mitigates the influence of view-specific and noisy view information on model performance. The proposed method outperforms numerous advanced approaches, as evidenced by extensive experiments conducted on eleven challenging multiview datasets. Particularly, it achieves 66.4%, 74.7%, 82.3%, and 86.4% clustering accuracies on the four Caltech datasets with different views, respectively.

关键词： Representation learning Accuracy Laplace equations data mining Contrastive learning Robustness Matrix decomposition Feature extraction data models Computational modeling Fusion contrastive learning (CL) fusion network (FNet) multilevel multiview clustering (MVC)

来源：评论

学校读者我要写书评

暂无评论

IntervenGen: Interventional data Generation for Robust and data-Efficient Robot Imitation learning

IntervenGen: Interventional Data Generation for Robust and D...

引用

2024 International conference on Intelligent Robots and systems

作者： Hoque, Ryan Mandlekar, Ajay Garrett, Caelan Goldberg, Ken Hox, Dieter Univ Calif Berkeley Berkeley CA 94720 USA NVIDIA Santa Clara CA 95051 USA

ISBN: (纸本)9798350377712;9798350377705

Imitation learning is a promising paradigm for training robot control policies, but these policies can suffer from distribution shift, where the conditions at evaluation time differ from those in the training data. A popular approach for increasing policy robustness to distribution shift is interactive imitation learning (i.e., DAgger and variants), where a human operator provides corrective interventions during policy rollouts. However, collecting a sufficient amount of interventions to cover the distribution of policy mistakes can be burdensome for human operators. We propose IntervenGen (I-Gen), a novel data augmentation system for robot control that autonomously produces a large set of corrective interventions with rich coverage of the state space from a small number of human interventions. We apply I-Gen to 4 simulated environments and 1 physical environment with object pose estimation error and show that it can increase policy robustness by up to 39x with only 10 human interventions. Videos and more results are available at https://***/view/intervengen2024.

关键词： Contrastive 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：