检索结果-内蒙古大学图书馆

Transfer Learning for EEG-Based Brain-Computer Interfaces: A Review of Progress Made Since 2016

IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS 2022年第1期14卷 4-19页

作者： Wu, Dongrui Xu, Yifan Lu, Bao-Liang Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Minist Educ Key Lab Image Proc & Intelligent Control Wuhan 430074 Peoples R China Shanghai Jiao Tong Univ Ctr Brain Comp & Machine Intelligence Key Lab Shanghai Educ Commiss Intelligent Interac Dept Comp Sci & EngnBrain Sci & Technol Res Ctr Shanghai 200240 Peoples R China Shanghai Jiao Tong Univ Qing Yuan Res Inst Shanghai 200240 Peoples R China

A brain-computer interface (BCI) enables a user to communicate with a computer directly using brain signals. The most common noninvasive BCI modality, electroencephalogram (EEG), is sensitive to noise/artifact and suffers between-subject/within-subject nonstationarity. Therefore, it is difficult to build a generic pattern recognition model in an EEG-based BCI system that is optimal for different subjects, during different sessions, for different devices and tasks. Usually, a calibration session is needed to collect some training data for a new subject, which is time consuming and user unfriendly. Transfer learning (TL), which utilizes data or knowledge from similar or relevant subjects/sessions/devices/tasks to facilitate learning for a new subject/session/device/task, is frequently used to reduce the amount of calibration effort. This article reviews journal publications on TL approaches in EEG-based BCIs in the last few years, i.e., since 2016. Six paradigms and applications-motor imagery, event-related potentials, steady-state visual evoked potentials, affective BCIs, regression problems, and adversarial attacks-are considered. For each paradigm/application, we group the TL approaches into cross-subject/session, cross-device, and cross-task settings and review them separately. Observations and conclusions are made at the end of the article, which may point to future research directions.

关键词： Electroencephalography Task analysis Calibration Brain modeling Machine learning Visualization Probability distribution Adversarial attacks affective brain-computer interface (BCI) brain-computer interfaces domain adaptation electroencephalogram (EEG) transfer learning (TL)

来源：评论

学校读者我要写书评

暂无评论

Impulsive Communication With Full and Partial Information for Adaptive Tracking Consensus of Uncertain Second-Order Multiagent Systems

引用

IEEE TRANSACTIONS ON CYBERNETICS 2022年第10期52卷 10302-10313页

作者： Han, Yiyan Zeng, Zhigang Huazhong Univ Sci & Technol Inst Artificial Intelligence Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Key Lab Image Proc & Intelligent Control Educ Minist China Wuhan 430074 Peoples R China

The uncertainty of dynamics is often unavoidable in practical applications especially for networked control systems while energy cost reduction is a perpetual issue for engineering. With this motivation, this article considers the adaptive tracking consensus problem of uncertain second-order multiagent systems via impulsive communication. The tracking consensus is achieved by designing proper adaptive control schemes with neural networks. Two control strategies are proposed for different cases. One considers that all state information is available while another considers only partial information can be used. Estimators equipped by followers are designed, which do not need any information about neighbors during the time interval without communication. Even though the estimators and followers are under control continuously over time, the communication among all agents is only permitted at impulsive instants. In such situations, some sufficient conditions to guarantee the estimation and convergence are obtained for both cases. It is proved that by the proposed adaptive schemes for uncertain multiagent systems, errors exist both for estimation and consensus due to uncertain dynamics and adaptive schemes. Numerical simulations, including a practical example, are presented to illustrate the effectiveness of the proposed method.

关键词： Multi-agent systems Protocols Integrated circuit modeling Estimation Uncertainty System performance Topology Adaptive tracking consensus estimator impulsive communication (IC) partial information uncertain second-order multiagent systems

来源：评论

学校读者我要写书评

暂无评论

Distributed Secondary control for Voltage Regulation and Optimal Power Sharing in DC Microgrids

引用

IEEE TRANSACTIONS ON control SYSTEMS TECHNOLOGY 2022年第6期30卷 2561-2572页

作者： Dou, Yangming Chi, Ming Liu, Zhi-Wei Wen, Guanghui Sun, Qihai Huazhong Univ Sci & Technol Key Lab Image Proc & Intelligent Control Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Minist Educ Wuhan 430074 Peoples R China Southeast Univ Sch Math Dept Syst Sci Nanjing 211189 Peoples R China

In this article, a novel distributed secondary control scheme is proposed to solve the voltage regulation and power sharing problems of an islanded direct current (DC) microgrid. It not only eliminates the voltage deviation caused by droop control but also minimizes the total generation cost of the DC microgrid. The proposed secondary control of distributed generators (DGs) does not require global information, but only current information of neighbors and the DC bus voltage. Different from the existing secondary frequency control, the corresponding closed-loop system of the DC microgrid under the proposed scheme is a nonlinear networked system. With the help of the properties of cascade systems and the input-to-state stability theory, it is found that the DC microgrid under the proposed scheme can achieve voltage regulation and minimization of the total generation cost if the corresponding communication graph of DGs is connected. A simulation model is established to verify the proposed scheme. Finally, a DC microgrid platform with two DGs in parallel is used for the experimental verification.

关键词： Microgrids Voltage control Costs Generators Power generation Resistance Resource management DC microgrids economic dispatch input-to-state stability power sharing voltage regulation

来源：评论

学校读者我要写书评

暂无评论

Energy management strategy based on health state for a PEMFC/Lithium-ion batteries hybrid power system

引用

ENERGY CONVERSION AND MANAGEMENT 2022年 271卷

作者： Sheng, Chuang Fu, Jun Li, Dong Jiang, Chang Guo, Ziang Li, Beijia Lei, Jingzhi Zeng, Linghong Deng, Zhonghua Fu, Xiaowei Li, Xi Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Key Lab Image Proc & Intelligent Control Educ Mini Wuhan 430074 Peoples R China Shenzhen Huazhong Univ Sci Technol Res Inst Shenzhen 518055 Peoples R China

The traditional energy management strategies (EMSs) for fuel cell/battery hybrid vehicle systems frequently neglect the degradation issues of power sources. The health-conscious EMS can take active control over the operation, thus mitigating the detrimental effects of degradation. Here, we will discuss an EMS that takes into account both proton exchange membrane fuel cell (PEMFC) and lithium-ion battery (LIB) degradation. To develop the desired strategy, performance degradation models for two power sources are constructed. It is the first time that the aggregation of platinum particles is taken into account in the water content model and used as the degradation index of PEMFC. The model's accuracy is checked by comparing the voltage drop when the model outputs steadily with data from experiments. Meanwhile, the optimal range of water content is deter-mined based on the analysis of the impact of water content on the stack voltage degradation. This provides a benchmark for the health control of PEMFC in EMS. Then, accurate knowledge of battery SoC can contribute to the rational management of LIBs. An improved UKF (unscented Kalman filter) algorithm is utilized to jointly estimate the SoC and model parameters of the battery and is compared with other methods. It can be found that the proposed joint estimation method has more precise SoC estimation results than the EKF (extended Kalman filter) and the traditional UKF approaches. This provides a reliable basis for the regulation of LIBs in EMS. Finally, taking PEMFC water content, battery SoC, and the required power as inputs, a rule-based health -conscious EMS is developed. The power response curve obtained from the HIL (hardware-in-the-loop) simulation platform shows that the proposed EMS can meet different load demand changes. In addition, during the test, the water content in the stack is kept in the range of 14-18.5, and the battery SoC is kept in the range of 0.6-0.8. Both of them are within the established benchma

关键词： PEMFC Lithium -ion battery Platinum particles agglomeration UKF Energy management strategy

来源：评论

学校读者我要写书评

暂无评论

Optimal guaranteed cost intermittent control to the efficient movement of freight trains

引用

ISA TRANSACTIONS 2022年 130卷 235-252页

作者： Hu, Junhao Ruan, Zeyu Mei, Jun Wang, Bo Ramatsetse, Boitumelo South Cent Minzu Univ Sch Math & Stat Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Minist Educ Image Proc & Intelligent Control Key Lab Wuhan 430074 Peoples R China Univ Witwatersrand Educ Informat & Engn Technol Johannesburg South Africa

Freight train system is under constant pressure to optimize operational efficiency, a factor that has led the sector to spearhead many new technologies. To address this, the optimal guaranteed cost intermittent cruise control of freight train system with time-varying uncertain parameters is investigated. In particular, the guaranteed cost intermittent control approach is proposed, which can achieve both advantages of guaranteed cost control and intermittent control methods. In addition, the acceptable conditions for the guaranteed cost control laws are weakened to insure that the proposed strategy is suitable for the application in freight train systems. Moreover, to promote the optimal control design in the freight train system, the proposed optimal guaranteed cost intermittent control for the nonlinear system subjected to norm bounded parametric uncertainties is addressed. The results of numerical experiments are presented to ascertain the effectiveness of the proposed freight train control methodology and to compare it to prior methods. (c) 2022 ISA. Published by Elsevier Ltd. All rights reserved.

关键词： Freight trains Guaranteed cost intermittent control Optimal control Energy efficient movement

来源：评论

学校读者我要写书评

暂无评论

Coarse-To-Fine Incremental Few-Shot Learning 1

引用

17th European Conference on Computer Vision (ECCV)

作者： Xiang, Xiang Tan, Yuwen Wan, Qian Ma, Jing Yuille, Alan Hager, Gregory D. Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Key Lab Image Proc & Intelligent Control Minist Educ Wuhan Peoples R China Johns Hopkins Univ Dept Comp Sci Baltimore MD 21218 USA

ISBN: (数字)9783031198212

ISBN: (纸本)9783031198205;9783031198212

Different from fine-tuning models pre-trained on a large-scale dataset of preset classes, class-incremental learning (CIL) aims to recognize novel classes over time without forgetting pre-trained classes. However, a given model will be challenged by test images with finer-grained classes, e.g., a basenji is at most recognized as a dog. Such images form a new training set (i.e., support set) so that the incremental model is hoped to recognize a basenji (i.e., query) as a basenji next time. This paper formulates such a hybrid natural problem of coarse-to-fine few-shot (C2FS) recognition as a CIL problem named C2FSCIL, and proposes a simple, effective, and theoretically-sound strategy Knowe: to learn, freeze, and normalize a classifier's weights from fine labels, once learning an embedding space contrastively from coarse labels. Besides, as CIL aims at a stability-plasticity balance, new overall performance metrics are proposed. In hat sense, on CIFAR-100, BREEDS, and tieredimageNet, Knowe outperforms all recent relevant CIL or FSCIL methods.

关键词： Class-incremental learning Coarse-to-fine Few shots

来源：评论

学校读者我要写书评

暂无评论

Dynamic Graph Dismantling with Reinforcement Supervised Learning

Dynamic Graph Dismantling with Reinforcement Supervised Lear...

引用

Chinese control Conference (CCC)

作者： Yumei Wang Chuancong Tang Hai-Tao Zhang the Engineering Research Center of Autonomous Intelligent Unmanned Systems the Key Laboratory of Image Processing and Intelligent Control and the State Key Laboratory of Digital Manufacturing Equipment and Technology School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China

There are always some “key” nodes in a big complex network, which can joint the most connected subgraphs. How to identify these nodes, finding a minimum set of nodes to attack for reducing the size of residual network's Largest Connected Component(LCC) to break up the original network, has become a research hotspot. Therefore, a method for determining the “key” nodes based on reinforcement learning framework and supervised learning model is proposed. This algorithm can not only utilize the dynamic exploration ability of reinforcement learning to collect a rich training dataset, but also take advantage of the characteristics that supervised learning is adaptive and has strong generalization ability to possess high efficiency and strong robustness. In order to further improve the algorithm's performance, $\epsilon$ -greedy mechanism is used to explore more network states. The experiment results show that given the same fraction of removed nodes, our algorithm can make the residual LCC smaller in various networks which is superior to the state-of-the-art algorithms in terms of effectiveness and generalization.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A new energy storage sharing framework with regard to both storage capacity and power capacity

引用

APPLIED ENERGY 2022年 307卷 118171-118171页

作者： Xiao, Jiang-Wen Yang, Yan-Bing Cui, Shichang Liu, Xiao-Kang Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Minist Educ Key Lab Image Proc & Intelligent Control Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Elect & Elect Engn Wuhan 430074 Peoples R China

Energy storage systems have received widespread attention due to their advantages on rapid response, smooth fluctuations, and the reduction of temporal and spatial imbalance. At present, most researchers mainly consider the allocation of energy storage capacity while using an average allocation of the power capacity, which may lead to conflicts among users when executing the energy sharing strategies for the case with limited power capacity. In order to better improve energy efficiency and reduce electricity costs, this paper proposes an energy storage sharing framework considering both the storage capacity and the power capacity. Since each prosumer intends to minimize its costs which depend on the behavior of other prosumers, the relationship among prosumers forms a non-cooperative game. By using an algorithm that combines the alternating direction multiplier method and heavy ball method, a generalized Nash equilibrium is found to serve as an energy sharing strategy. Simulation studies and comparisons show that the proposed energy storage sharing framework driven by a dynamic electricity price mechanism can reduce prosumers' net demands, electricity costs and improve the efficiency of energy storage utilization.

关键词： Community energy storage sharing Prosumer Dynamic electricity price mechanism Non-cooperative game Distributed optimization Heavy ball method

来源：评论

学校读者我要写书评

暂无评论

USV Path Planning Based on ACO-SA Algorithm

USV Path Planning Based on ACO-SA Algorithm

引用

Chinese control Conference (CCC)

作者： Chuanshang Luo Yifeng Tang Zhihui Huang Yifan Xu Bin Liu Hai-Tao Zhang the Engineering Research Center of Autonomous Intelligent Unmanned Systems the Key Laboratory of Image Processing and Intelligent Control and the State Key Laboratory of Digital Manufacturing Equipment and Technology School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China

This paper solves USV path planning problem constrained by multiple factors via ant-colony optimization algorithm. First, this paper uses the ways of multi-objective optimization to model the USV path planning problem. An improved ant colony algorithm named ACO-SA is put forward afterwards to effectively solve the problem. The algorithm is a combination of ACO algorithm(ant colony algorithm) and SA algorithm(simulated annealing algorithm), which has three improments: change the initial distribution of pheromone to guide the search when the algorithm has just started running; change the heuristic function and state transition probability taking three factors into consideration; change the pheromone update rule and make the ants compete for the right to update pheromone by simulated annealing algorithm, and update the best solution by the same algorithm. Finally, simulation experiment and field experiment are conducted to check the validity of ACO-SA algorithm.

关键词：

来源：评论

学校读者我要写书评

暂无评论

3D Instances as 1D Kernels 1

引用

17th European Conference on Computer Vision (ECCV)

作者： Wu, Yizheng Shi, Min Du, Shuaiyuan Lu, Hao Cao, Zhiguo Zhong, Weicai Huazhong Univ Sci & Technol Minist Educ Sch AIA Key Lab Image Proc & Intelligent Control Wuhan Peoples R China Huawei CBG Consumer Cloud Serv Search & Maps BU Shenzhen Peoples R China

ISBN: (数字)9783031198182

ISBN: (纸本)9783031198175;9783031198182

We introduce a 3D instance representation, termed instance kernels, where instances are represented by one-dimensional vectors that encode the semantic, positional, and shape information of 3D instances. We show that instance kernels enable easy mask inference by simply scanning kernels over the entire scenes, avoiding the heavy reliance on proposals or heuristic clustering algorithms in standard 3D instance segmentation pipelines. The idea of instance kernel is inspired by recent success of dynamic convolutions in 2D/3D instance segmentation. However, we find it non-trivial to represent 3D instances due to the disordered and unstructured nature of point cloud data, e.g., poor instance localization can significantly degrade instance representation. To remedy this, we construct a novel 3D instance encoding paradigm. First, potential instance centroids are localized as candidates. Then, a candidate merging scheme is devised to simultaneously aggregate duplicated candidates and collect context around the merged centroids to form the instance kernels. Once instance kernels are available, instance masks can be reconstructed via dynamic convolutions whose weights are conditioned on instance kernels. The whole pipeline is instantiated with a dynamic kernel network (DKNet). Results show that DKNet outperforms the state of the arts on both ScanNetV2 and S3DIS datasets with better instance localization.

关键词： Instance kernel Point cloud Instance segmentation

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：