检索结果-内蒙古大学图书馆

UAV-Based Real-Time Air Quality Monitoring and Prediction Using Machine Learning Approach

IEEE Internet of Things Journal 2025年

作者： Hsu, Huan-Chia Tarekegn, Getaneh Berie Getnet, Mastewal Endeshaw Yu, Kuo-Pin Tai, Li-Chia NYCU Department of Electrical and Computer Engineering Hsinchu30010 Taiwan NYCU Institute of Electrical and Control Engineering Hsinchu30010 Taiwan NYCU Institute of Environmental and Occupational Health Sciences Taipei112304 Taiwan NYCU Department of International Health Program Taipei112304 Taiwan

Globally, air pollution has become a major problem due to the growth of the human population and rapid urbanization, posing a serious threat to the environment and human health. Currently, most monitoring systems are based on fixed monitoring stations and are inefficient due to slow response time. An accurate and real-time air quality monitoring and prediction (AQMP) system is required to prevent health problems caused by air pollution. With the swift advancement of Internet of Things (IoT) technologies such as low-cost sensors and advanced wireless communication systems capable of sensing their environment, connecting with other systems, and interacting with users. To overcome the problems of existing systems, this paper aims to develop an unmanned aerial vehicle (UAV)-based real-time AQMP system by integrating artificial intelligence and data science technologies. The proposed UAV is equipped with lightweight air quality sensors including particulate matter (i.e., PM2.5 and PM10), carbon monoxide (CO), carbon dioxide (CO2), and ozone (O3), and meteorological sensors. The sensors are mounted on the UAV via the microcontroller, and the sensors reading data is transmitted to the web server via a message queueing telemetry transport (MQTT) protocol using a 5G mobile network module. Practical experiments were conducted in Hsinchu City and the industrial area of Yunlin County to collect real-time air pollutant data. Then, we employed a hybrid of attention mechanism, convolutional neural network (CNN) and gated recurrent unit (GRU) algorithms to predict future air quality levels. Experimental results show that the proposed system improves prediction performance compared with baseline time series algorithms. © 2014 IEEE.

关键词： Air quality

来源：评论

学校读者我要写书评

暂无评论

Geo2ComMap: Deep Learning-Based MIMO Throughput Prediction Using Geographic Data

引用

IEEE Wireless Communications Letters 2025年

作者： Lin, Fan-Hao Huang, Tzu-Hao Wen, Chao-Kai Duong, Trung Q. National Sun Yat-sen University Institute of Communications Engineering Kaohsiung80424 Taiwan National Yang Ming Chiao Tung University Institute of Electrical Control Engineering Hsinchu300 Taiwan Memorial University Faculty of Engineering and Applied Science St. John'sNLA1C 5S7 Canada Queen's University Belfast School of Electronics Electrical Engineering and Computer Science Belfast United Kingdom

Accurate communication performance prediction is crucial for wireless applications such as network deployment and resource management. Unlike conventional systems with a single transceiver antenna, throughput (Tput) estimation in antenna array-based multiple-output multiple-input (MIMO) systems is computationally intensive, i.e., requiring analysis of channel matrices, rank conditions, and spatial channel quality. These calculations impose significant computational and time burdens. This paper introduces Geo2ComMap, a deep learning-based framework that leverages geographic databases to efficiently estimate multiple communication metrics across an entire area in MIMO systems using only sparse measurements. To mitigate extreme prediction errors, we propose a sparse sampling strategy. Extensive evaluations demonstrate that Geo2ComMap accurately predicts full-area communication metrics, achieving a median absolute error of 27.35 Mbps for Tput values ranging from 0 to 1900 Mbps. © 2012 IEEE.

关键词： Antenna arrays

来源：评论

学校读者我要写书评

暂无评论

Interpretable deep temporal neural networks for in-situ monitoring under varying conditions in micro-electrical discharge machining

引用

Journal of Manufacturing Systems 2025年 81卷 222-237页

作者： Long Ye Cheng Guo Ming Wu Jun Qian Nan Yu Dominiek Reynaerts Institute of Materials and Processes School of Engineering University of Edinburgh Edinburgh EH9 3FB UK Guangdong Provincial Key Laboratory of Micro/Nano Optomechatronics Engineering College of Mechatronics and Control Engineering Shenzhen University Shenzhen 518060 PR China Micro & Precision Engineering Group Department of Mechanical Engineering KU Leuven Leuven 3001 Belgium Members Flanders Make Leuven Belgium Department of Computer Science KU Leuven Leuven 3001 Belgium

Monitoring is critical enabler of digitalization in modern manufacturing, supporting enhanced process control, quality assurance, and real-time decision-making. By integrating data-driven techniques with the powerful capabilities of deep learning, monitoring systems can efficiently extract valuable insights from complex, high-dimensional time-series data. However, traditional data-driven approaches often lack interpretability, limiting their adoption in industrial applications that demand high reliability and accountability. To address this challenge, this paper proposes an interpretable monitoring framework based on a deep temporal neural network (DTNN). Designed with a modular architecture, the DTNN integrates key components for embedding, temporal feature learning and classification, enabling it to effectively capture complex underlying patterns in temporal process data and overcome the limitations of conventional methods. The DTNN’s capabilities are demonstrated in the context of micro-electrical discharge machining (micro-EDM), a prominent non-traditional machining process known for producing intricate and high-precision components. Through a pulse discrimination task utilizing a large dataset of reliable labels, the DTNN achieves superior classification accuracy under varying processing parameters while providing interpretable insights into discharge phenomena. Furthermore, the DTNN monitoring approach is applied to a deep-hole drilling process in micro-EDM, enabling closed-loop control of discharge status and ensuring long-term process stability. The DTNN’s modular design, interpretability and real-time adaptability underscore its potential for advancing data-driven monitoring systems in digital manufacturing.

关键词：

来源：评论

学校读者我要写书评

暂无评论

DEEPCON: Improving Distributed Deep Learning Model Consistency in Edge-Cloud Environments via Distillation

引用

IEEE Transactions on Cognitive Communications and Networking 2025年

作者： Qian, Bin Qian, Jiaxu Wen, Zhenyu Wu, Di He, Shibo Chen, Jiming Ranjan, Rajiv Zhejiang University State Key Laboratory of Industrial Control Technology Hangzhou China Zhejiang University of Technology Institute of Cyberspace Security College of Information Engineering Hangzhou China University of St Andrews School of Computer Science United Kingdom Newcastle University United Kingdom

In a typical distributed Deep Learning (DL) based application, models are configured differently to meet the requirements of resource constraints. For instance, a large ResNet56 model is deployed on the cloud server while a small lightweight MobileNet model is more suitable for the end-user device with fewer computation resources. However, the heterogeneity of the model architectures and configurations may bring a systemic problem -models may produce different outputs when given the same input. This inconsistency problem may cause severe system failure of prediction agreement inside the application. Current research has not studied the systemic design for efficiently detecting and reducing the inconsistency among models in distributed DL applications. With the increasing scale of distributed DL applications, the challenges of inconsistency mitigation should consider both algorithm and system design. To this end, we design and implement DEEPCON, an adaptive deployment system across the edge-cloud layer with over-the-air model updates. We implement ASRS sampling for efficiently sampling data to reveal the real data distribution as well as model prediction inconsistency. Then, we implement DMML-Par, an asynchronous parallel training algorithm for quickly updating the models and reducing inconsistency. implements over-the-air updates with a set of APIS to enable seamless inconsistency detection and reduction in such deep learning applications. Our experiment results on both vision and language tasks demonstrate that DMML could improve the model consistency up to 4%, 7%, and 13% at CIFAR10/100 and IMDB datasets without sacrificing the accuracy of individual models. We also show that the ASRS sampling can save 90% network bandwidth of data transmission and that DMML-Par is up to 60% faster compared to simple synchronous parallel training. 2015 IEEE.

关键词： Distributed cloud

来源：评论

学校读者我要写书评

暂无评论

Symmetry in linear physical systems

arXiv

引用

arXiv 2025年

作者： van der Schaft, Arjan Sepulchre, Rodolphe Chaffey, Thomas Bernoulli Institute for Mathematics Computer Science and Artificial Intelligence Jan C. Willems Center for Systems and Control University of Groningen PO Box 407 9700 AK Netherlands KU Leuven Belgium University of Cambridge United Kingdom School of Electrical and Computer Engineering University of Sydney Australia

Physical systems with symmetry arise abundantly in applications, and are endowed with interesting mathematical structures. The present paper focusses on linear reciprocal and input-output Hamiltonian systems. Their characterization is studied from an input-output as well as from a state point of view. Geometrically, it turns out that they both define Lagrangian subspaces with corresponding generating functionals. Furthermore, the relations with time reversibility are analyzed. The system classes under consideration are expected to admit scalable control laws, and to be important building blocks in design. Copyright © 2025, The Authors. All rights reserved.

关键词： Hamiltonians

来源：评论

学校读者我要写书评

暂无评论

Distributed Prescribed-Time Observer for Nonlinear Systems

arXiv

引用

arXiv 2025年

作者： de Heij, Vincent Niazi, M.B. Umar Johansson, Karl H. Ahmed, Saeed Faculty of Science and Engineering University of Groningen Groningen9747 AG Netherlands Division of Decision and Control Systems Digital Futures KTH Royal Institute of Technology StockholmSE-100 44 Sweden Laboratory for Information and Decision Systems Department of Electrical Engineering and Computer Science Massachusetts Institute of Technology CambridgeMA02139 United States

This paper proposes a distributed prescribed-time observer for nonlinear systems representable in a block-triangular observable canonical form. Using a weighted average of neighbor estimates exchanged over a strongly connected digraph, each observer estimates the system state despite limited local sensor measurements. The proposed design guarantees that distributed state estimation errors converge to zero at a user-specified convergence time, irrespective of observers’ initial conditions. To achieve this prescribed-time convergence, distributed observers implement time-varying local output injection gains that asymptotically approach infinity as the prescribed time is approached. The theoretical convergence is rigorously proven and validated through numerical simulations. Copyright © 2025, The Authors. All rights reserved.

关键词： Nonlinear simulations

来源：评论

学校读者我要写书评

暂无评论

QoI-Aware Configuration Adaptation and Heterogeneous Resource Allocation for Edge Video Analytics

引用

IEEE Internet of Things Journal 2025年

作者： He, Zixuan Sun, Yanjing Zhang, Beibei Dong, Kaiwen Wang, Bowen Xu, Hongli China University of Mining and Technology School of Information and Control Engineering Xuzhou221116 China China University of Mining and Technology IoT Perception Mine Research Center Xuzhou221116 China School of Information and Control Engineering China University of Science and Technology of China School of Computer Science and Technology Anhui Hefei230027 China University of Science and Technology of China Suzhou Institute for Advanced Research Jiangsu Suzhou215123 China

Edge video analytics enables agile responses of machine-centric applications by streaming videos from end devices to edge servers (ESs) for resource-intensive Deep Neural Network (DNN) inference. Quality of Inference (QoI), reflected by end-to-end analytics delay and inference accuracy, is crucial for the high-quality real-time decision-making. Due to the limited ES resources and dynamic network conditions, video configuration adaptation and fine-grained resource allocation are necessary to enhance the QoI of video streams, which involves a delicate balance between delay and accuracy. In this paper, we propose an edge-enabled multi-video analytics framework, in which performing DNN inference relies on heterogeneous resources, including CPU, GPU, and memory. Considering the different impacts of heterogeneous resources on QoI and double-queue backlogs, we formulate the joint video configuration adaptation, CPU-GPU resource allocation, and batch size selection (JCCGB) problem based on semi-Markov Decision Process to maximize the long-term average QoI. Then, Lyapunov optimization is applied to transform the original problem into one that minimizes the Lyapunov drift-plus-penalty upper bound, thereby ensuring the stability of both the transmission and computation queues. To tackle the potential multimodality of the optimal configuration adaptation and resource allocation policy, we propose the diffusion deep reinforcement learning based JCCGB (DD-JCCGB) scheme. This approach effectively enhances the policy performance and accelerates the training process. Experiments driven by real-world network traces demonstrate that the DD-JCCGB algorithm improves the long-term average QoI and outperforms baseline schemes. © 2025 IEEE.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

A Recurrent Neural Network Based Bicriteria Repetitive Motion Collision Avoidance Scheme for Motion Planning of Dual Redundant Manipulators 13

A Recurrent Neural Network Based Bicriteria Repetitive Motio...

引用

13th International Conference on Intelligent control and Information Processing, ICICIP 2025

作者： Chen, Shirui Gao, Huanli Hu, Shaolin Zhang, Zhijun The School of Automation Science and Engineering South China University of Technology Guangzhou510640 China The School of Automation Guangdong University of Petrochemical Technology Maoming Guangzhou525000 China The Guangzhou Key Laboratory of Brain-Computer Interaction and Applications Guangzhou China The Key Library of Autonomous Systems and Network Control Ministry of Education The School of Automation Science and Engineering South China University of Technology Guangzhou510640 China Guangzhou510555 China The Jiangxi Thousand Talents Plan Nanchang University Nanchang330031 China The College of Computer Science and Engineering Jishou University Jishou416000 China Guangzhou510335 China The Shaanxi Provincial Key Laboratory of Industrial Automation School of Mechanical Engineering Shaanxi University of Technology Hanzhong723001 China The School of Information Science and Engineering Changsha Normal University Changsha410100 China The School of Automation Science and Engineering Maoming525000 China The Institute of Artificial Intelligence and Automation Guangdong University of Petrochemical Technology Maoming525000 China

ISBN: (纸本)9798331516147

To achieve low joint-angle drift and avoid mutual collision between dual redundant manipulators (DRMs) when they are doing collaboration works, a recurrent neural network based bicriteria repetitive motion collision avoidance (BRMCA) scheme is proposed for motion planning of DRMs. By designing a combined optimization objective encompassing both repetitive motion criterion and collision avoidance criterion, the BRMCA scheme has larger feasible region and less constraint contradiction issues compared with traditional inequality constraints based method. Furthermore, by incorporating velocity-level kinematic model and physical limits of DRMs into the scheme, a time varying quadratic programming (TVQP) problem is formulated. The linear variational inequality-based primal–dual neural network (LVI-PDNN) is used to solve the proposed scheme due to the high accuracy and low computation complexity. Finally, through comparison simulations of tracking Circle-shape and C-shape trajectories between BRMCA scheme and minimum velocity norm (MVN) scheme, it exhibits that the proposed BRMCA scheme can achieve low joint-angle drift, collision avoidance and high tracking accuracy when DRMs are doing cooperative tasks. © 2025 IEEE.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Climate Downscaling Using Neural Operator: Spatiotemporal Multimodal Fusion Operator with State-Query Coupled Kernel

Climate Downscaling Using Neural Operator: Spatiotemporal Mu...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Haodi Zhang Yichi Wang Yifan Jian Jiahui Jiang Zhaohai Bai Lin Ma College of Computer Science and Software Engineering Shenzhen University Center for Agricultural Resources Research Institute of Genetic and Developmental Biology Chinese Academy of Sciences State Key Laboratory of Pollution Control and Resource Reuse School of the Environment Nanjing University

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Climate downscaling is crucial for detailed small- scale analysis and for acquiring climate data in regions without weather stations. Operator learning has proven potential for this task. However, several challenges remain in operator learning, such as multimodal fusion, spatiotemporal fusion and input state and query adaptation. To address these challenges, we propose a Spatiotemporal Multimodal Fusion Operator with a State- Query Coupled Kernel (SMCK). This framework includes a latent space fusion encoder that encodes climate variables using position-wise multihead attention for multimodal fusion and integrates historical information to generate robust and precise representation. Additionally, we introduce a state-query coupled kernel that combines radial basis functions and discrete fourier encoding to enhance query location representation, while also adapting to the state to obtain the coupled kernel. Extensive experiments demonstrate that our method achieves state-of-the-art performance and provides strong support for climate downscaling and the planning of climate-related strategies.

关键词： Couplings Signal processing Encoding Robustness Spatiotemporal phenomena Planning Kernel Speech processing Forecasting Meteorology

来源：评论

学校读者我要写书评

暂无评论

Robust Jumping of A Bipedal Wheel Legged Robot in Gusty Weather 24

Robust Jumping of A Bipedal Wheel Legged Robot in Gusty Weat...

引用

Proceedings of the 2024 6th International Conference on Mechatronics Systems and control engineering

作者： Yiyu Guo Minghui Zhang Qihong Guo Jingyan Sun School of Mechanical Engineering Southwest Jiaotong University Chengdu Sichuan China Institute of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing Beijing China School of Mechanical Engineering Fuzhou University Fuzhou Fujian China School of Control and Computer Engineering North China Electric Power University Beijing Beijing China

ISBN: (纸本)9798400710346

Currently, bipedal wheeled-leg robots are considered to have significant potential for various applications. Current research lacks an in-depth study of the performance of bipedal wheeled-leg robots in high-wind conditions. However, researching robots in such conditions is crucial for their application in outdoor environments, such as disaster rescue. Therefore, enhancing the balance performance of these robots to improve their mobility in complex environments is crucial. This paper explores the jumping performance of bipedal wheeled-leg robots in strong wind conditions and proposes a method for optimizing the robot's jumping trajectory using a genetic algorithm. This approach enables the robot to successfully perform jumps and maximize the jump distance even in windy environments. To make the simulation environment more realistic, a detailed simulation of strong wind conditions was conducted, where a function was established to model the variation in wind force on the robot's surface under different tilt angles during a jump. Additionally, the maximum torque limitations of the motors and the weight of the electromechanical components were taken into account. Ultimately, successful jumps were achieved in windy conditions, with a great improvement in jump distance compared to before optimization.

关键词： Bipedal wheeled-leg robot

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：