检索结果-内蒙古大学图书馆

A Real-Time Semantic Segmentation Method Based on Transformer for Autonomous Driving

computers, Materials & Continua 2024年第12期81卷 4419-4433页

作者： Weiyu Hao Jingyi Wang Huimin Lu Donald Bren School of Information and Computer Sciences University of CaliforniaIrvineCA 92612USA Department of Control Engineering Kyushu Institute of TechnologyKitakyushu804-8550Japan School of Automation Southeast UniversityNanjing210096China

While traditional Convolutional Neural Network(CNN)-based semantic segmentation methods have proven effective,they often encounter significant computational challenges due to the requirement for dense pixel-level predictions,which complicates real-time *** address this,we introduce an advanced real-time semantic segmentation strategy specifically designed for autonomous driving,utilizing the capabilities of Visual *** leveraging the self-attention mechanism inherent in Visual Transformers,our method enhances global contextual awareness,refining the representation of each pixel in relation to the overall *** enhancement is critical for quickly and accurately interpreting the complex elements within driving sce-narios—a fundamental need for autonomous *** experiments conducted on the DriveSeg autonomous driving dataset indicate that our model surpasses traditional segmentation methods,achieving a significant 4.5%improvement in Mean Intersection over Union(mIoU)while maintaining real-time *** paper not only underscores the potential for optimized semantic segmentation but also establishes a promising direction for real-time processing in autonomous navigation *** work will focus on integrating this technique with other perception modules in autonomous driving to further improve the robustness and efficiency of self-driving perception frameworks,thereby opening new pathways for research and practical applications in scenarios requiring rapid and precise decision-making *** experimentation and adaptation of this model could lead to broader implications for the fields of machine learning and computer vision,particularly in enhancing the interaction between automated systems and their dynamic environments.

关键词： Visual transformer semantic segmentation autonomous driving

来源：评论

学校读者我要写书评

暂无评论

Incremental Weighted Ensemble for Data Streams With Concept Drift

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第1期5卷 92-103页

作者： Jiao, Botao Guo, Yinan Yang, Cuie Pu, Jiayang Zheng, Zhiji Gong, Dunwei School of Electromechanical and Information Engineering Beijing100083 China China University of Mining and Technology School of Information and Control Engineering Xuzhou221116 China Victoria University of Wellington School of Engineering and Computer Wellington6140 New Zealand Electromechanical and Information Engineering Beijing100083 China Qingdao University of Science and Technology School of Information Science and Technology Qingdao266061 China

As a popular strategy to tackle concept drift, chunk-based ensemble method adapts a new concept by adjusting the weights of historical classifiers. However, most previous approaches normally evaluate the historical classifier based on an entire chunk newly arrived, which may cause delayed adaptation. To address the issue, two novel ensemble models, named incremental weighted ensemble (IWE) and incremental weighted ensemble for multi-classification (IWE-M), are proposed. At each time step, all base classifiers are incrementally updated on a newly arrived instance. Following that, the instance is collected into a cache array. Once a data chunk is formed, a new base classifier is created. More specially, a forgetting mechanism based on variable-size window is designed to adjust the weight of each base classifier in IWE in terms of its classification accuracy on the latest instances in an online manner. IWE-M, an extension of IWE, aims to solve multiclass problems with local concept drifts. In IWE-M, the weight of a base classifier is expanded to a weight vector. In this way, this ensemble model can retain specific historical information about nondrift regions from a local drift. Experimental results show that the proposed ensemble frameworks outperform six competitive approaches on accuracy and G-mean. © 2020 IEEE.

关键词： Classifiers

来源：评论

学校读者我要写书评

暂无评论

AFSTGCN:Prediction for multivariate time series using an adaptive fused spatial-temporal graph convolutional network

引用

Digital Communications and Networks 2024年第2期10卷 292-303页

作者： Yuteng Xiao Kaijian Xia Hongsheng Yin Yu-Dong Zhang Zhenjiang Qian Zhaoyang Liu Yuehan Liang Xiaodan Li School of Information and Control Engineering China University of Mining and TechnologyXuzhou221116China Department of Informatics University of LeicesterLeicesterLE17RHUK The Affiliated Changshu Hospital of Soochow University Changshu215500China School of Computer Science and Engineering Changshu Institute of TechnologySuzhou215500China

The prediction for Multivariate Time Series(MTS)explores the interrelationships among variables at historical moments,extracts their relevant characteristics,and is widely used in finance,weather,complex industries and other ***,it is important to construct a digital twin ***,existing methods do not take full advantage of the potential properties of variables,which results in poor predicted *** this paper,we propose the Adaptive Fused Spatial-Temporal Graph Convolutional Network(AFSTGCN).First,to address the problem of the unknown spatial-temporal structure,we construct the Adaptive Fused Spatial-Temporal Graph(AFSTG)***,we fuse the spatial-temporal graph based on the interrelationship of spatial ***,we construct the adaptive adjacency matrix of the spatial-temporal graph using node embedding ***,to overcome the insufficient extraction of disordered correlation features,we construct the Adaptive Fused Spatial-Temporal Graph Convolutional(AFSTGC)*** module forces the reordering of disordered temporal,spatial and spatial-temporal dependencies into rule-like *** dynamically and synchronously acquires potential temporal,spatial and spatial-temporal correlations,thereby fully extracting rich hierarchical feature information to enhance the predicted *** on different types of MTS datasets demonstrate that the model achieves state-of-the-art single-step and multi-step performance compared with eight other deep learning models.

关键词： Adaptive adjacency matrix Digital twin Graph convolutional network Multivariate time series prediction Spatial-temporal graph

来源：评论

学校读者我要写书评

暂无评论

Graph Dual Adversarial Network for Hyperspectral Image Classification

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2023年第4期4卷 922-932页

作者： Cheng, Yuhu Chen, Yang Kong, Yi Philip Chen, C.L. Wang, Xuesong China University of Mining and Technology Engineering Research Center of Intelligent Control for Underground Space Ministry of EducationSchool of Information and Control Engineering Xuzhou221116 China South China University of Technology School of Computer Science and Engineering Guangzhou510006 China

An end-to-end unsupervised domain adaptation method for hyperspectral image (HSI) classification based on a graph dual adversarial network is proposed in this article. First, in order to extract the domain-invariant features of the source and target domains, the rich spectral information and spatial position of HSI are used to construct a spectral-spatial nearest neighbor graph, which is input into the graph convolutional network. Then, a prototype adversarial strategy is proposed, which uses the labeled data of the source domain to reliably calculate the feature prototypes of different classes. Through the prototype adversarial strategy, the distances between different prototypes are appropriately extended, so that the clusters of different classes are far away from their respective decision boundaries, and the discriminability of features are also enhanced. The dual adversarial strategy is composed of the prototype adversarial strategy and the domain adversarial strategy. It is worth noting that the dual adversarial strategy does not require feature extractor and discriminator to work in turn, which can be implemented through a gradient reversal layer. Finally, on the basis of adapting the overall features of both the domains via the domain adversarial strategy, the source- and target-domain features are further adapted by minimizing the correlation alignment loss of each class of samples. Experimental results on two real HSI datasets of Botswana and Kennedy Space Center show the effectiveness of our proposed method. © 2020 IEEE.

关键词： Spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Prescribed-Time Nash Equilibrium Seeking for Pursuit-Evasion Game

引用

IEEE/CAA Journal of Automatica Sinica 2024年第6期11卷 1518-1520页

作者： Lei Xue Jianfeng Ye Yongbao Wu Jian Liu D.C.Wunsch Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of EducationNanjing 210096 School of Automation Southeast UniversityNanjing 210096China Department of Electrical and Computer Engineering Missouri University of Science and TechnologyRollaMO 65409 USA IEEE

Dear Editor,This letter is concerned with prescribed-time Nash equilibrium(PTNE)seeking problem in a pursuit-evasion game(PEG)involving agents with second-order *** order to achieve the prior-given and user-defined convergence time for the PEG,a PTNE seeking algorithm has been developed to facilitate collaboration among multiple pursuers for capturing the evader without the need for any global ***,it is theoretically proved that the prescribedtime convergence of the designed algorithm for achieving Nash equilibrium of ***,the effectiveness of the PTNE method was validated by numerical simulation results.A PEG consists of two groups of agents:evaders and *** pursuers aim to capture the evaders through cooperative efforts,while the evaders strive to evade *** is a classic noncooperative *** has attracted plenty of attention due to its wide application scenarios,such as smart grids[1],formation control[2],[3],and spacecraft rendezvous[4].It is noteworthy that most previous research on seeking the Nash equilibrium of the game,where no agent has an incentive to change its actions,has focused on asymptotic and exponential convergence[5]-[7].

关键词： seeking prescribed convergence

来源：评论

学校读者我要写书评

暂无评论

A Physical-Data Fusion Framework for Lithium-ion Battery SOC Estimation∗ 49

A Physical-Data Fusion Framework for Lithium-ion Battery SOC...

引用

49th Annual Conference of the IEEE Industrial Electronics Society, IECON 2023

作者： Fu, Bingzhe Li, Yihuan Liu, Kailong Lin, Wei School of Control and Computer Engineering North China Electric Power University Beijing China School of Control Science and Engineering Shandong University Shandong China

ISBN: (纸本)9798350331820

Lithium-ion batteries have increasingly become a primary energy source in Electric Vehicles (EVs), power grid energy storage, aerospace, and other fields. Accurate State of Charge (SOC) estimation is crucial for the safe and efficient operation of lithium batteries. This paper proposes a physical-data fusion framework for accurate SOC estimation of Lithium-ion batteries. First, a fractional-order model (FOM) of lithium-ion batteries is built to describe the electrochemical reactions inside the battery, and a fractional-order Extended Kalman Filter (FOEKF) algorithm is used to achieve preliminary SOC estimation. Second, the FOM is combined with an error model established by Long Short-Term Memory (LSTM) to improve the accuracy of the FOEK-based SOC estimation by compensating for the estimation error. Finally, the proposed framework is verified on Maryland dataset, and the experimental results demonstrate that the fusion framework exhibits superior performance in improving the accuracy of SOC estimation, with at least 75% and 50% reduction in MSE and max error, respectively. © 2023 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Fourier Series Fitting-Based Intrusion Detection for Real-Time Industrial control Systems

Fourier Series Fitting-Based Intrusion Detection for Real-Ti...

引用

2024 International Conference on Artificial Intelligence of Things and Systems, AIoTSys 2024

作者： Zhang, Heng Wang, Yinchu Xue, Yawen Deng, Ruilong Li, Hongran Zhang, Jian School of Computer Engineering Jiangsu Ocean University Lianyungang China College of Control Science and Engineering Zhejiang University Hangzhou China

ISBN: (纸本)9798331507886

Recent incidents have shown the attacks against Industrial control System (ICS), which may result in failure or even catastrophe and prompt the studies of defense. Existing security systems are mostly based on network perimeter defenses which lacking effective countermeasures in the event that such defenses are bypassed and attackers gain access to the system. In this paper, we propose a Fourier series fitting based intrusion detection method, which enables a detection and an alarm when ICS is under attack. Specifically, we first divide the data into three parts based on time with a sliding window, namely real-Time data, sliding window data to be predicted, and historical data. Then we fit the data within the prediction window to the data around the window, together with the root mean squared error (RMSE) calculated for the data within the window and the historical data. The intrusion detection system (IDS) will alarm when the difference between the above RMSE and the reference one is greater than a threshold value. We test two types of data injection attacks, extensive simulations illustrate the effectiveness of our method in detecting both of these attacks. © 2024 IEEE.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Character recognition competition for street view shop signs

引用

National Science Review 2023年第6期10卷 27-31页

作者： Jingqun Tang Weidong Du Bin Wang Wenyang Zhou Shuqi Mei Tao Xue Xing Xu Hai Zhang Shanghai Byte Dance Technology Co. Ltd School of Control Science and Engineering Zhejiang University School of Computer Science and Technology University of Science and Technology of China School of Computer Science and Technology Tsinghua Univeristy Shenzhen Tencent Technology Co. Ltd School of Computer Science and Engineering University of Electronic Science and Technology of China Pazhou Laboratory (Huangpu) School of Mathematics Northwest University

INTRODUCTION Store signboards provide important information in street view images,andcharacter recognition in natural scenes is an important research direction in computer *** view store signboard character recognitio...

关键词：

来源：评论

学校读者我要写书评

暂无评论

Improving Spiking Neural Network Performance Using Astrocyte Feedback for Farsi Digit Recognition 31

Improving Spiking Neural Network Performance Using Astrocyte...

引用

31st International Conference on Electrical engineering, ICEE 2023

作者： Nazari, Malihe Bahrami, Fariba Yazdanpanah, Mohammad Javad School of Electrical and Computer Engineering College of Engineering University of Tehran Human Motor Control and Computational Neuroscience Lab Advanced Control Systems Lab Tehran Iran School of Electrical and Computer Engineering College of Engineering University of Tehran Human Motor Control and Computational Neuroscience Lab Tehran Iran School of Electrical and Computer Engineering College of Engineering University of Tehran Advanced Control Systems Lab Tehran Iran

ISBN: (纸本)9798350312560

The third generation of neural networks is called spiking neural networks. Spiking neural networks can not only answer all the problems that can be solved by common neural networks, they can also be computationally more powerful than them. Also, these networks are closer to the biological reality of neurons. According to these reasons, spiking neural networks have received much attention in recent years. In addition to neurons, other elements are also used in spiking neural networks, whose connection with neurons has been proven biologically. These elements are called astrocytes. In this study, we have used astrocyte local feedback in a spiking neural network to improve the accuracy of the network in a digit recognition process. To reach this goal, we use the network proposed by Diehl et al in 2015, then we added astrocyte local feedback to this network. We run the network in both states. The classification accuracy of the network in the absence of astrocytes was 88.48% on Farsi handwritten digits. After adding astrocyte local feedback, it is observed that astrocytes control the activity of neurons and modified the weight of connections based on neuron activity. In this state classification accuracy reached 90.26%. The idea was to create a neural network for learning, where, inspired by the physiological role of astrocytes at the tripartite synapse improves the performance of the neural network and we demonstrate that systems that include neurons and astrocytes perform better than systems incorporating only neurons. © 2023 IEEE.

关键词： Neurons

来源：评论

学校读者我要写书评

暂无评论

A novel fractional physics-informed neural networks method for solving the time-fractional Huxley equation

引用

Neural Computing and Applications 2024年第30期36卷 19097-19119页

作者： Shi, Jieyu Yang, Xiaozhong Liu, Xinlong School of Control and Computer Engineering North China Electric Power University Beijing102206 China Institute of Information and Computing School of Mathematics and Physics North China Electric Power University Beijing102206 China

The neural network methods in solving differential equations have significant research importance and promising application prospects. Aimed at the time-fractional Huxley (TFH) equation, we propose a novel fractional physics-informed neural networks (fPINNs) method. By integrating the physical information of the TFH equation into neural networks, the fPINNs are trained as a precise approximation model for solving the TFH equation. The fPINNs method involves calculating the Caputo fractional derivative using the L1 formula and estimating the integer-order derivative through the chain rule. The Adam algorithm is employed to optimize two kinds of loss functions constructed by hard and soft constraints, respectively. Through an analysis of the impact of fPINNs parameters on training results, we identify the optimal parameter configuration for solving both one-dimensional and two-dimensional (2D) TFH equations. Numerical examples validate the efficiency and robustness of the fPINNs method in solving TFH equations. Furthermore, numerical simulation of the 2D TFH problem demonstrates the method’s practical applicability to engineering problem. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Numerical methods

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：