检索结果-内蒙古大学图书馆

IEEE International Conference on Big Data

作者： Danilo Menegatti Alessandro Giuseppi Cecilia Poli Antonio Pietrabissa Dept. of Computer Control and Management Engineering Sapienza University of Rome Rome Italy National Centre for Innovative Technologies in Public Health Department Istituto Superiore di Sanità Rome Italy

ISBN: (数字)9798350362480

ISBN: (纸本)9798350362497

Topology optimization in decentralised federated learning settings enables the design of policies aimed at minimizing the number of communication rounds needed to reach algorithmic convergence. Given a federation of autonomous agents, finding the optimal topology which guarantees that the underlying graph is connected is still an open issue. This paper proposes a novel energy-aware topology optimization algorithm with the goal to derive an optimal topology which maximizes the algebraic connectivity of the corresponding graph in presence of energy and communication constraints. The effectiveness of the proposed approach is validated in the context of a consensus-based federated learning algorithm over an e-Health scenario.

关键词： Federated learning Heuristic algorithms Big Data Autonomous agents Topology Optimization Convergence

来源：评论

学校读者我要写书评

暂无评论

NLOS/LOS Classification by Constructing Indirect Environment Interaction From GNSS Receiver Measurements Using a Transformer-Based Deep Learning Model

NLOS/LOS Classification by Constructing Indirect Environment...

引用

2023 International Technical Meeting of The Institute of Navigation, ITM 2023

作者： Li, Zhenni Zeng, Kungan Wang, Lujia Kan, Xie Yuan, Rong Xie, Shengli School of Automation Guangdong University of Technology Guangzhou510006 China Guangdong-HongKong-Macao Joint Laboratory for Smart Discrete Manufacturing Guangzhou510006 China School of Software and Engineering Sun Yat-sen University Hong Kong Department of Electronic and Computer Engineering Hong Kong University of Science and Technology Hong Kong Techtotop Microelectronics Technology Co. Ltd Guangzhou510006 China Guangdong Key Laboratory of IoT Information Technology Guangzhou510006 China

ISBN: (纸本)9781713871378

The blockage of the direct signal by high-rise buildings degrades the positioning performance of the global navigation satellite system (GNSS) in urban canyons. Traditional methods usually distinguish the line-of-sight (LOS), and non-line-of-sight (NLOS) signals using the satellite’s elevation angle or signal strength to mitigate the issue, which cannot obtain satisfactory classification accuracy in complex urban environments. Recently, machine learning-based methods using GNSS measurements have been developed to improve GNSS signal classification. However, dynamic changes in environments and the limited features from the measurements raise a challenge to the classification performance of algorithms. To address the above issues, a transformer-based deep learning model in this paper is proposed for LOS and NLOS signal classification. The innovation of our proposed model is using the self-attention mechanism to construct the indirect environment interaction for learning the representation of environmental information. First, the measurements of captured satellites in the same epoch will be fed into the model. The model’s feature extraction layer can learn the fundamental representation. Then, the self-attention module can construct the indirect environment interaction by computing the relevance between the satellites. Finally, the module’s output is transferred to the output layer for classification. A comparative experimental study is conducted with real-world data collected in urban canyons. The results demonstrate that our model can keep more than 90% classification accuracy while support vector machine (SVM) and multilayer perception (MLP) decline to about 80% with the sites increasing. © 2023 Proceedings of the International Technical Meeting of The Institute of Navigation, ITM.

关键词： Satellites

来源：评论

学校读者我要写书评

暂无评论

A Cloud-Edge-Based Multi-Label Learning and Fast Clustering Method for Emerging New PV Faults

A Cloud-Edge-Based Multi-Label Learning and Fast Clustering ...

引用

IEEE International Conference on Cyber Technology in automation, control, and Intelligent Systems

作者： Qi Liu Bo Yang Yuxiang Liu Kai Ma Xinping Guan Department of Automation Ministry of Education Key Laboratory of System Control and Information Processing Shanghai Jiao Tong University Shanghai China School of Electrical Engineering Yanshan University Qinhuangdao China

Aiming at accurate identification of emerging new faults in PV systems, a novel PV fault diagnosis framework based on cloud-edge collaboration is designed. Model updates are deployed on the cloud servers, while real-time diagnosis is implemented at the edge devices. A multi-label learning model is developed to extract fault features efficiently, and a threshold scheme is further designed for feature enhancement. The obtained features are processed by K-Means clustering and the One-Class Support Vector Machine (OCSVM) methods, which can accurately diagnose existing types and identify unknown faults. For analysis and demonstration, the Principal Component Analysis (PCA) and t-Distributed Stochastic Neighbor Embedding (t-SNE) methods are used for dimensionality reduction and visualization. Numerical results validate the effectiveness of the proposed method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

3DOS: towards 3D open set learning - benchmarking and understanding semantic novelty detection on point clouds 22

3DOS: towards 3D open set learning - benchmarking and unders...

引用

Proceedings of the 36th International Conference on Neural Information Processing Systems

作者： Antonio Alliegro Francesco Cappio Borlino Tatiana Tommasi Department of Control and Computer Engineering. Politecnico di Torino Italy and Italian Institute of Technology Italy

ISBN: (纸本)9781713871088

In recent years there has been significant progress in the field of 3D learning on classification, detection and segmentation problems. The vast majority of the existing studies focus on canonical closed-set conditions, neglecting the intrinsic open nature of the real-world. This limits the abilities of robots and autonomous systems involved in safety-critical applications that require managing novel and unknown signals. In this context exploiting 3D data can be a valuable asset since it provides rich information about the geometry of perceived objects and scenes. With this paper we provide the first broad study on 3D Open Set learning. We introduce 3DOS: a novel testbed for semantic novelty detection that considers several settings with increasing difficulties in terms of semantic (category) shift, and covers both in-domain (synthetic-to-synthetic, real-to-real) and cross-domain (synthetic-to-real) scenarios. Moreover, we investigate the related 2D Open Set literature to understand if and how its recent improvements are effective on 3D data. Our extensive benchmark positions several algorithms in the same coherent picture, revealing their strengths and limitations. The results of our analysis may serve as a reliable foothold for future tailored 3D Open Set methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Wear Attenuation in Actuators based on a Model-free Adaptive Filter for a Gantry Crane System

Wear Attenuation in Actuators based on a Model-free Adaptive...

引用

control Technology and Applications (CCTA),

作者： Lucas M. Gomes José Oniram de A. Limaverde Filho Eugênio Fortaleza Andres F. Gonzalez Fernando P. Munerato Department of Mechanical Engineering Automation and Control Group (GRACO) University of Brasília Brasília Brazil Research and Development Repsol Sinopec Brasil Rio de Janeiro RJ Brazil

Gantry cranes systems are widely used in heavy load transportation industries, which increasingly requires engineering solutions to improve efficiency while considering safety precautions. Aiming to reduce the actuator wear of these systems, this article proposes to add a model-free adaptive filter in the control loop to mitigate the impact of measurement noise on the control signal without compromising the overall closed-loop system performance. The main idea is to avoid the sudden change of the filtered signal when the system output has a derivative not statistically significant when compared to the expected measurement noise. Numerical simulation and experimental validation are performed in a gantry crane system to demonstrate the effectiveness and practicability of the adaptive filter.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An Approach for Stock Price Prediction Based on Improved Eemd and Attention-Enhanced Bilstm

SSRN

引用

SSRN 2024年

作者： Liu, Zhinzhong Su, Zhaohui Shang, Ling Sun, Hongxiang Zhao, Bin School of Computer and Control Engineering Yantai University Yantai264005 China Digital Smart Creative Department Henan Culture and Tourism Investment Group Co. Ltd Luoyang471026 China

Accurate stock price prediction is a urgently needed technology in economic investment. However, stock prices are influenced by various factors and display non-linear and non-stationary characteristics, which makes accurate prediction of stock prices a critical challenge. To tackle the above issue, we propose a method for stock price prediction based on improved ensemble empirical mode decomposition (EEMD) and attention-enhanced BiLSTM (AEBiLSTM)(named as SPPEA). First, to reduce the nonlinearity of the stock prices and improve the prediction accuracy, our method first divides the original stock price series into sub-sequences with the same time scale using a sliding window. These sub-sequences are then decomposed into multiple intrinsic mode Function (IMF) components with the EEMD method which is improved by cubic spline interpolation. Then, the decomposed IMF components are input into AEBiLSTM to capture the dependencies of stock prices data and the features of important points of the stock prices, and then the prediction results of each IMF component are obtained with AEBiLSTM. Finally, the prediction results of each component are integrated to obtain the final prediction results. Extensive experiments have been conducted on three stock datasets of BYD, China Unicom and Ping An Bank, experimental results prove that our proposed method has better performance compared with the state-of-the-art methods. © 2024, The Authors. All rights reserved.

关键词： Empirical mode decomposition

来源：评论

学校读者我要写书评

暂无评论

Comprehensive electrical models for a wireless sensor network device

引用

Heliyon 2024年第23期10卷 e40415页

作者： León Ávila, Bernardo Yaser García Vázquez, Carlos Alberto Pérez Baluja, Osmel Cotfas, Daniel Tudor Cotfas, Petru Adrian Transilvania University of Brașov Department of Electronics and Computers Faculty of Electrical Engineering and Computer Science Brasov 500036 Romania University of Sancti Spiritus “José Martí Pérez” Department of Computer Engineering Faculty of Business and Technical Sciences Comandante Manuel Fajardo s/n Olivos 1 Sancti Spíritus 60100 Cuba Technological University of Havana Department of Automation and Computing Faculty of Automation and Biomedical Engineering Street 114 No. 11901 between Ciclovía and Rotonda Havana 19390 Cuba

This paper proposes a novel circuit model of a Wireless Sensor Network (WSN) device. The model is designed to accurately represent the behavior of a WSN device operating in a duty cycle, capturing the essential characteristics of its components, including the microcontroller, sensor, transceiver, and power supply. The proposed model incorporates static and dynamic power consumption aspects, reflecting the energy usage patterns during various operational states such as sensing, processing, and communication. The model's accuracy is validated through a comparative analysis with a LTspice® implementation and a real mote implemented for this study. Simulation results closely match the empirical data, demonstrating the model's effectiveness in predicting the mote's behavior. This approach provides a valuable tool for optimizing power consumption and extending the lifetime of WSN deployments within a simulation environment, offering significant insights for designers and researchers in this field. © 2024 The Authors

关键词： Electric model SPICE model Wireless sensor network node

来源：评论

学校读者我要写书评

暂无评论

Self‑Generated Buried Submicrocavities for High‑Performance Near‑Infrared Perovskite Light‑Emitting Diode

引用

Nano-Micro Letters 2023年第8期15卷 355-367页

作者： Jiong Li Chenghao Duan Qianpeng Zhang Chang Chen Qiaoyun Wen Minchao Qin Christopher C.S.Chan Shibing Zou Jianwu Wei Zuo Xiao Chuantian Zuo Xinhui Lu Kam Sing Wong Zhiyong Fan Keyou Yan School of Environment and Energy State Key Lab of Luminescent Materials and DevicesGuangdong Provincial Key Laboratory of Solid Wastes Pollution Control and RecyclingSouth China University of TechnologyGuangzhou 510000People’s Republic of China Department of Electronic and Computer Engineering The Hong Kong University of Science and TechnologyClear Water BayKowloonHong KongPeople’s Republic of China Department of Physics The Chinese University of Hong KongShatinHong KongPeople’s Republic of China Department of Physics The Hong Kong University of Science and TechnologyClear Water BayKowloon 999077Hong KongPeople’s Republic of China School of Chemistry and Chemical Engineering Guangxi UniversityNanning 530004People’s Republic of China Center for Excellence in Nanoscience(CAS) Key Laboratory of Nanosystem and Hierarchical Fabrication(CAS)National Center for Nanoscience and TechnologyBeijing 100190People’s Republic of China

ABSTRACT Embedding submicrocavities is an effective approach to improve the light out-coupling efficiency(LOCE)for planar perovskite light-emitting diodes(PeLEDs).In this work,we employ phenethylammonium iodide(PEAI)to trigger the Ostwald ripening for the downward recrystallization of perovskite,resulting in spontaneous formation of buried submicrocavities as light output *** simulation suggests the buried submicrocavities can improve the LOCE from 26.8 to 36.2%for near-infrared ***,PeLED yields peak external quantum efficiency(EQE)increasing from 17.3%at current density of 114 mA cm^(−2)to 25.5%at current density of 109 mA cm^(−2)and a radiance increasing from 109 to 487 W sr^(−1)m^(−2)with low *** turn-on voltage decreased from 1.25 to 1.15 V at 0.1 W sr^(−1)m^(−2).Besides,downward recrystallization process slightly reduces the trap density from 8.90×10^(15)to 7.27×10^(15)cm^(−3).This work provides a self-assembly method to integrate buried output coupler for boosting the performance of PeLEDs.

关键词： Perovskite light-emitting diodes Downward recrystallization Buried submicrocavities Light out-coupling efficiency Radiative recombination

来源：评论

学校读者我要写书评

暂无评论

Risk-Aware Non-Myopic Motion Planner for Large-Scale Robotic Swarm Using CVaR Constraints

arXiv

引用

arXiv 2024年

作者： Yang, Xuru Hu, Yunze Gao, Han Ding, Kang Li, Zhaoyang Zhu, Pingping Sun, Ying Liu, Chang Department of Advanced Manufacturing and Robotics College of Engineering Peking University Beijing100871 China Department of Automation Tsinghua University Beijing100080 China Marshall University HuntingtonWV25755 United States School of Electrical Engineering and Computer Science The Pennsylvania State University State CollegePA16802 United States

Swarm robotics has garnered significant attention due to its ability to accomplish elaborate and synchronized tasks. Existing methodologies for motion planning of swarm robotic systems mainly encounter difficulties in scalability and safety guarantee. To address these limitations, we propose a Risk-aware swarm mOtion planner using conditional ValuE-at-Risk (ROVER) that systematically navigates large-scale swarms through cluttered environments while ensuring safety. ROVER formulates a finite-time model predictive control (FTMPC) problem predicated upon the macroscopic state of the robot swarm represented by a Gaussian Mixture Model (GMM) and integrates conditional value-at-risk (CVaR) to ensure collision avoidance. The key component of ROVER is imposing a CVaR constraint on the distribution of the Signed Distance Function between the swarm GMM and obstacles in the FTMPC to enforce collision avoidance. Utilizing the analytical expression of CVaR of a GMM derived in this work, we develop a computationally efficient solution to solve the non-linear constrained FTMPC through sequential linear programming. Simulations and comparisons with representative benchmark approaches demonstrate the effectiveness of ROVER in flexibility, scalability, and safety guarantee. Copyright © 2024, The Authors. All rights reserved.

关键词： Predictive control systems

来源：评论

学校读者我要写书评

暂无评论

Domain generalized solar cell defect segmentation based on shape-aware and multi-view meta-learning

Domain generalized solar cell defect segmentation based on s...

引用

第42届中国控制会议

作者： Kun Liu Ying Yang Jingkun Mao Weipeng Liu School of Artificial Intelligence and Data Science Hebei University of Technology Control Engineering Technology Innovation Center of Hebei Province Hebei University of Technology Department of Automation Tianjin University of Technology

Various defects are inevitably generated in the manufacturing process of solar *** learning-based methods for defect segmentation under closed situation have achieved remarkable *** to the difference of imaging condition and camera parameter under different production line,there are large differences in brightness distribution of solar cell *** model trained under closed situation cannot achieve good performance in opened situation such as multi-production *** this paper,a new shape-aware and multi-view meta-learning scheme is proposed to improve the model generalization performance for single-domain solar cell defect *** scheme roots in gradient-based *** view information is yielded from the generated augmented images,and explicitly simulating domain shift with virtual meta-train and meta-test during training to mitigate overfitting in single-source domain train and unstable ***,considering tiny and faint solar cell defects are not easily identified under different image brightness conditions,shape edge constraint loss is proposed to encourage shape compactness and shape smoothness of *** results show that the proposed method has high segmentation accuracy,which outperforms the state-of-the-art methods.

关键词：

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：