检索结果-内蒙古大学图书馆

2nd ieee International conference on Integrated Intelligence and Communication systems, ICIICS 2024

作者： Mishra, Soumya Ranjan Dash, Sachikanta Rath, Lingaraj Department of CSA GIET University Gunupur India Department of CSE GIET University Gunupur India

ISBN: (纸本)9798331504960

Demand for healthcare Internet of things (IoT) devices that forecast and track illnesses has been driven by the rising worldwide prevalence of diabetes, which impacts more than 346 million people. A healthcare approach that does not involve invasive IoT monitoring and uses machine learning to forecast the likelihood of diabetes from key indicators like blood glucose levels is proposed in this study. The major objective is to create a program that assists diabetics in keeping their condition under control. Predicting the consequences of diabetes, the algorithm achieves 98.41 percent accuracy by using machine learning techniques, particularly bagging and boosting. For the purpose of training the model, 10,221 users' data was culled from the PIMA database (PID), questionnaires, and IoT sensors. The IoT app provided a superior solution for predicting diabetes risk variables by integrating three conventional machine learning methods with a hybrid ensemble approach. This allowed it to outperform previous models. © 2024 ieee.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Automated Vehicle Identification Based on Car-Following data With Machine learning

引用

ieee TRANSACTIONS ON INTELLIGENT TRANSPORTATION systems 2023年第12期24卷 13893-13902页

作者： Li, Qianwen Li, Xiaopeng Yao, Handong Liang, Zhaohui Xie, Weijun Univ Georgia Sch Environm Civil Agr & Mech Engn Athens GA 30602 USA Univ Wisconsin Madison Dept Civil & Environm Engn Madison WI 53706 USA Georgia Tech H Milton Stewart Sch Ind & Syst Engn Atlanta GA 30332 USA

Vehicles with adaptive cruise control, i.e., SAE Levels 1 and 2 automated vehicles (AVs), have been operating on roads with a significant and rapidly growing penetration rate. Identifying these AVs is critical to understanding near-future mixed traffic characteristics and managing highway mobility and safety. This study identifies adaptive cruise control-equipped vehicles from human-driven vehicles (HVs) by constructing a set of learning-based models using car-following trajectories in a short time window. It is extendible to Level 3 and $+$ AV identification when data is available. To compare model performance and draw physical insights, two physics-based models are proposed based on the premise that, in general, the car-following behavior of an AV is less volatile than an HV. Four car-following datasets, including AV makes from different manufacturers, are mixed to build a comprehensive identification model. Results show that physics-based approaches identify more than 80% AVs and 70% HVs. The identification accuracy of learning-based models is even higher. For example, the cluster-aware long short-term memory network identifies 98.79% of AVs and 95.45% of HVs. learning-based identification models developed by this study can be integrated with the existing infrastructure (e.g., surveillance cameras), which have been used to extract car-following trajectories, to detect AVs in mixed traffic streams. This opens unparalleled data-driven opportunities to analyze and control mixed traffic to enhance safety (e.g., notifying surrounding traffic of the presence of AVs) and mobility (e.g., opening AV dedicated lanes when the percentage is great enough).

关键词： Roads Safety Trajectory Global Positioning System data models Behavioral sciences Accidents Automated vehicle identification adaptive cruise control machine learning model physics-based model car following

来源：评论

学校读者我要写书评

暂无评论

From data-driven to Model-driven learning via Structured Dynamic Mode Decomposition 63

From Data-Driven to Model-Driven Learning via Structured Dyn...

引用

63rd ieee conference on Decision and control, CDC 2024

作者： Bonenberger, Christopher Scholz, Stephan Schneider, Markus University of Applied Sciences Ravensburg-Weingarten Weingarten88250 Germany Ulm University Ulm89081 Germany Weingarten88250 Germany Weingarten88250 Germany

ISBN: (纸本)9798350316339

We consider the modeling process of physics-informed Dynamic Mode Decomposition (piDMD), proposing a novel formalism that allows to formulate physical principles in a detailed fashion. More precisely, we constrain the solution space by explicitly specifying the system to be estimated as a sum of elementary matrices with desired properties. Moreover, we augment this approach towards DMD with control and provide closed-form solutions for a wide class of shift-equivariant systems. Finally, the proposed approach opens an alternative perspective on piDMD that also allows for efficient implementations. © 2024 ieee.

关键词： Dynamic mode decomposition

来源：评论

学校读者我要写书评

暂无评论

data-driven Anti-Disturbance control with High-Order Extended State Observer for Unknown Nonlinear systems 14

Data-Driven Anti-Disturbance Control with High-Order Extende...

引用

14th Asian control conference (ASCC)

作者： Yue, Shuhao Hao, Shoulin Liu, Tao Yu, Haichen Zhang, Jiyan Wang, Haixia Dalian Univ Technol Inst Adv Control Technol Dalian Peoples R China Dalian Univ Technol Sch Control Sci & Engn Dalian Peoples R China

ISBN: (纸本)9798331540845;9789887581598

For nonlinear systems with unknown dynamics subject to external disturbances, a data-driven anti-disturbance control scheme is proposed by using only the measured system input and output data. A modified local compact-form dynamic linearization (CFDL) model is firstly established for the considered system. Then, a predictive anti-disturbance control law is designed by solving a constrained optimization problem in time domain. To make the control law implementable, the pseudo partial derivative (PPD) of the above model at the future time step is estimated by leveraging an auto-regressive prediction algorithm with the past PPD estimations. The residual disturbance to be predicted is constructed by using the current-time disturbance and its high-order difference estimated by an extended state observer (ESO). The convergence of the PPD estimation, ESO and the tracking error is analyzed with proofs. Finally, an illustrative example is given to validate the effectiveness and advantages of the proposed design over the existing methods.

关键词： Unknown nonlinear systems anti-disturbance control local compact-form dynamic linearization high-order extended state observer

来源：评论

学校读者我要写书评

暂无评论

Unbalanced Fault Classification Using Active learning in Synthetic Fiber Manufacturing Process 18

Unbalanced Fault Classification Using Active Learning in Syn...

引用

18th Annual ieee International systems conference (SysCon)

作者： Das, Shemonto Simha, Vimal Swamidas, Joshua Soares, Amilcar da Fonseca, Vinicius Prado Mem Univ Newfoundland Dept Comp Sci St John NF Canada Instrumar Ltd St John NF Canada Linnaeus Univ Dept Comp Sci & Media Technol Vaxjo Sweden

ISBN: (纸本)9798350358810;9798350358803

Fiber manufacturing is a complex and dynamic process where quality control poses unique challenges, requiring timely detection and accurate classification of faults. Automating the quality control process using machine-learning techniques has the potential to reduce costs and increase efficiency. However, training machine-learning models to classify faults requires labeling large quantities of time-series data generated from the sensors, which can be costly and time-consuming. In this work, we developed and evaluated Active learning (AL) techniques to enhance the performance of fault classification models while reducing the number of instances that require labeling. AL is based on focusing the labeling effort on a small number of selected, informative samples, thereby reducing the cost and time requirements for creating a labeled dataset. Moreover, in manufacturing, the frequencies of different types of faults vary widely, resulting in a class imbalance problem. The selection of instances per class to be labeled is beyond the control of traditional AL techniques, which may lead to significant bias in the classification performance and the selection of instances to be labeled for AL strategies, particularly in a dynamic manufacturing environment. To address this problem, we also develop and incorporate a class-balancing instance selection algorithm that tries to select more instances from the classes with fewer labeled examples. The AL techniques implemented here reduce the amount of labeled data necessary for accurate fault classification by a factor of 5 compared to conventional supervised machine-learning techniques. Additionally, our class-balancing instance selection algorithm effectively addresses the class imbalance problem in our dataset. Overall, our results indicate that our AL pipeline constitutes a promising solution for efficient and accurate fault classification using time-series data from industrial manufacturing.

关键词： Active learning Unbalanced data Fiber Manufacturing Process Fault Classification dataset Labeling

来源：评论

学校读者我要写书评

暂无评论

Graph-model Based Optimization of Shelter Hospitals: Integrating MILP for Efficient Layout and Deployment 13

Graph-model Based Optimization of Shelter Hospitals: Integra...

引用

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

作者： Daomeng, Cai Jinyuan, Liu Ertai, E. Tianyue, Wang Haoran, Xu Jinhui, Pang Yongling, Fu Beihang University No. 37 Xueyuan Road Haidian District Beijing China Systems Engineering Institute China State Shipbuilding Corporation No. 1 Fengxian East Road Haidian District Beijing China Beijing Institute of Technology No. 5 Zhongguancun South Street Haidian District Beijing China

ISBN: (纸本)9798350361674

This paper introduces a novel approach to optimizing the layout and deployment of mobile shelter hospitals, a concept that has gained prominence due to the increasing need for rapid medical response in diverse scenarios, including natural disasters and pandemics. Our approach integrates Mixed Integer Linear Programming (MILP) and the genetic algorithms to develop a graph model that represents the different functional areas of a shelter hospital as vertices. In this model, network edges are defined by capacity and transit time, employing a discrete time approach for patient movement and queuing. We establish a Nash equilibrium for the model to ensure optimal patient flow and treatment efficiency. Utilizing the genetic algorithms, the model's hyperparameters are refined, and we demonstrate its practical application through a case study. The results from our experiments indicate a significant improvement in layout planning and deployment strategies. In one specific instance, by optimizing the graph model parameters, the total time for all patients to receive services was reduced from 190 minutes to 170 minutes, a 10.5% reduction. This approach addresses the challenges of diverse and complex medical scenarios in shelter hospitals, enhancing their emergency response capabilities and operational efficiency. © 2024 ieee.

关键词： Mixed-integer linear programming

来源：评论

学校读者我要写书评

暂无评论

Reinforcement learning control for Autonomous Hydraulic Material Handling Machines with Underactuated Tools

Reinforcement Learning Control for Autonomous Hydraulic Mate...

引用

2024 International conference on Intelligent Robots and systems

作者： Spinelli, Filippo A. Egli, Pascal Nubert, Julian Nan, Fang Bleumer, Thilo Goegler, Patrick Brockes, Stephan Hofmann, Ferdinand Hutler, Marco Swiss Fed Inst Technol Robot Syst Lab Zurich Switzerland MPI Intelligent Syst Stuttgart Germany Liebherr Hydraulikbagger GmbH Kirchdorf Germany

ISBN: (纸本)9798350377712;9798350377705

The precise and safe control of heavy material handling machines presents numerous challenges due to the hard-to-model hydraulically actuated joints and the need for collision-free trajectory planning with a free-swinging end-effector tool. In this work, we propose an RL-based controller that commands the cabin joint and the arm simultaneously. It is trained in a simulation combining data-driven modeling techniques with first-principles modeling. On the one hand, we employ a neural network model to capture the highly nonlinear dynamics of the upper carriage turn hydraulic motor, incorporating explicit pressure prediction to handle delays better. On the other hand, we model the arm as velocity-controllable and the free-swinging end-effector tool as a damped pendulum using first principles. This combined model enhances our simulation environment, enabling the training of RL controllers that can be directly transferred to the real machine. Designed to reach steady-state Cartesian targets, the RL controller learns to leverage the hydraulic dynamics to improve accuracy, maintain high speeds, and minimize end-effector tool oscillations. Our controller, tested on a mid-size prototype material handler, is more accurate than an inexperienced operator and causes fewer tool oscillations. It demonstrates competitive performance even compared to an experienced professional driver.

关键词： End effectors

来源：评论

学校读者我要写书评

暂无评论

Machine learning for the control and Monitoring of Electric Machine Drives: Advances and Trends

IEEE OPEN JOURNAL OF INDUSTRY APPLICATIONS

引用

ieee OPEN JOURNAL OF INDUSTRY APPLICATIONS 2023年 4卷 188-214页

作者： Zhang, Shen Wallscheid, Oliver Porrmann, Mario Joby Aviat San Carlos CA 94070 USA Paderborn Univ Dept Power Elect & Elect Drives D-33098 Paderborn Germany Osnabruck Univ Inst Comp Sci D-49090 Osnabruck Germany

This review article systematically summarizes the existing literature on utilizing machine learning (ML) techniques for the control and monitoring of electric machine drives. It is anticipated that with the rapid progress in learning algorithms and specialized embedded hardware platforms, ML-based data-driven approaches will become standard tools for the automated high-performance control and monitoring of electric drives. In addition, this article also provides some outlook toward promoting its widespread application in the industry with a focus on deploying ML algorithms onto embedded system-on-chip field-programmable gate array devices.

关键词： Artificial intelligence (AI) deep learning electric machine drives embedded systems field-programmable gate array (FPGA) machine learning (ML) power electronics reinforcement learning (RL)

来源：评论

学校读者我要写书评

暂无评论

Hybrid data-driven and Physical Model for Distribution System State Estimation 4

Hybrid Data-Driven and Physical Model for Distribution Syste...

引用

4th International conference on Smart Grid and Energy Internet, SGEI 2024

作者： Xie, Biao Wang, Zhenyu Lu, Fang Song, Ziwei College of Intelligent Systems Science and Engineering Harbin Engineering University Heilongjiang Province Harbin China

ISBN: (纸本)9798350368314

Realizing accurate Distribution System State Estimation (DSSE) faces challenges including a lack of observability and high complexity of the distribution system. A data-driven approach based on machine learning models may be an option to address this issue, but due to the lack of data labels in practical applications, their application in DSSE is limited. In fact, measurements in distribution systems often contain noise, are damaged or missing, and are unusable. In order to better address these issues, a distribution system state estimation solver is proposed, which is a deep learning model based on Residual Network (ResNet) that study the relationship between measurement values and state variables. A learning method combining traditional machine learning and power flow guidance was proposed to train ResNet, which respects the physical laws of the distribution system by inputting the output of ResNet into the power flow equation. This strategy can learn from noisy measurements, reducing the need for ideal data labels. Experiments on ieee 33 bus distribution networks have shown that the model proposed in this paper outperforms traditional weighted least squares algorithms in accuracy. ©2024 ieee.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Clustered Predictive Quality of Service for Connected Vehicles Using Federated learning 59

Clustered Predictive Quality of Service for Connected Vehicl...

引用

59th Annual ieee International conference on Communications (ieee ICC)

作者： Stenhammar, Oscar Fodor, Gabor Fischione, Carlo KTH Royal Inst Technol Stockholm Sweden Ericsson Res Stockholm Sweden

ISBN: (纸本)9798350304060;9798350304053

To enable the development of multifunctional cellular networks that aim to satisfy increasing expectations of connectivity and trustworthiness, it is crucial to provide reliable quality of service (QoS) guarantees for end users. With predictive QoS (pQoS), cellular networks become proactive to meet QoS requirements for a wide variety of new use cases, including advanced driver assistance applications. This work introduces a novel predictive framework to improve the availability and performance of pQoS in cellular networks, especially for advanced road transport applications. We show that by dividing the road into geographical segments, clustering segments with similar data, and assigning each cluster a predictive model, the adversary effects of the propagation environment and interference on QoS become manageable. To this end, each predictive cluster model is trained locally on vehicles within the cluster boundaries by data driven Federated learning, resulting in personalized predictive models for each cluster. Our numerical results show that the clustered predictive model outperforms the more common predictive approach proposed by previous works that train a single global predictive model for an entire dataset.

关键词： 6G Quality of Service Deep learning Federated learning Vehicular and wireless technologies Advanced driver assistance systems

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：