检索结果-内蒙古大学图书馆

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

作者： Liu, Junhan Xu, Degang Guo, Yuqian School of Automation Central South University Changsha410083 China

ISBN: (纸本)9798350361674

To achieve accurate force control during slag removal operations in the metallurgical industry, the impedance control strategy is commonly employed. A robot designed for slag removal operation has been developed to handle oxidized slag suspended on the surface of molten metal. However, the oxidized slag typically adheres to the end-effector over time, altering the end-effector's weight and impacting the force control's precision. This paper proposes a novel force control strategy to realize the re-identification of variable parameters of the end-effector without stopping the operation of the robot. First, a six-dimensional force sensor is installed between the robot flange and the end-effector. The paper calibrates and calculates the end-effector's parameters, reducing the influence of the end-effector's gravity on force perception. Secondly, a real-time parameter updating algorithm, termed MRLS (Modified Recursive Least Squares Method), is designed specifically for parameter re-identification of variable robot payload. Finally, experimental results demonstrate that the error of the measured weight of the tool to the true value is 1.64%. When the robot is in the slag removal movement process, the error of the contact force estimated is within 0.13N. © 2024 ieee.

关键词： Slags

来源：评论

学校读者我要写书评

暂无评论

Guest Editorial of the Special Section on Neural Computing-driven Artificial Intelligence for Consumer Electronics

引用

ieee TRANSACTIONS ON CONSUMER ELECTRONICS 2024年第1期70卷 3517-3520页

作者： Zhang, Haijun Gao, Xiao-Zhi Wang, Zenghui Wang, Guanghui Harbin Inst Technol Dept Comp Sci Shenzhen 518055 Peoples R China Univ Eastern Finland Sch Comp Kuopio 70210 Finland Univ South Africa Dept Elect & Min Engn ZA-1709 Florida South Africa Toronto Metropolitan Univ Dept Comp Sci Toronto ON M5B 2K3 Canada

Recent advances in artificial intelligence (AI) technologies have driven the dramatic developments in key consumer applications, e.g., smart manufacturing, equipment conditions and fault diagnosis, quality inspection, autonomous decision-making, etc. In the Industry 4.0 era, AI has become the core technology to promote the revolution and development of consumer electronics intelligence. In practice, AI-driven consumer electronics integrate AI technologies and the domain knowledge of standard process and operations to achieve smart systems incorporated with techniques of the Internet of Things (IoT), neural computing, machine learning, and deep learning. However, many challenges are remained to implement AI-powered modes for consumer electronics by directly applying advanced neural computing techniques. Moreover, complex application context in consumer electronics environments and prior domain knowledge further make it challengeable to fulfill emerging intelligent consumer applications. On the other hand, recent years have witnessed the rapid development of neural computing in various AI tasks. In particular, deep neural networks have been widely applied in real-world application scenarios in consumer electronics manufacturing. Moreover, advanced techniques and approaches in data modeling and prediction, learning strategies, optimization and control theories are also incorporated and developed under various consumer application scenarios.

关键词： Special issues and sections Neural engineering Artificial intelligence Smart manufacturing Autonomous systems Decision making Smart devices Consumer electronics Internet of Things Fourth Industrial Revolution Fault diagnosis

来源：评论

学校读者我要写书评

暂无评论

Battery State-of-Charge Estimator Design based on The Least-Squares Support Vector Machine 20

Battery State-of-Charge Estimator Design based on The Least-...

引用

ieee 20th International conference on Automation Science and Engineering (CASE)

作者： Cavanini, Luca Majecki, Pawel Grimble, Mike J. van der Molen, Gerrit M. Ind Syst & Control Ltd Glasgow Lanark Scotland Univ Politecn Marche Ancona Italy Univ Strathclyde Glasgow Lanark Scotland

ISBN: (纸本)9798350358513;9798350358520

The design and development of a data-driven algorithm for battery State-of-Charge estimation is presented. The estimation of battery SoC is important in the development of Battery Management systems. The proposed approach exploits the Least-Squares Support Vector Machine data-driven estimation paradigm and statistical methods. The algorithm's computational complexity is reduced by using a data pruning procedure. The optimization of the SVM-based estimator is performed by using a Particle Swarm Optimization method. The design approach proposed to develop to estimator is validated using a simulation model of the battery and an Estimator Design Tool in MATLAB software which provides a user-friendly interface for the different algorithms that may be used in the estimator design. The approach is applicable to a wide range of applications including automotive systems.

关键词： Battery Management System Least-Squares Support Vector Machine State-of-Charge Estimator

来源：评论

学校读者我要写书评

暂无评论

data-driven explicit predictive control with limited resources: an exploration-based strategy 20

Data-driven explicit predictive control with limited resourc...

引用

20th IFAC Symposium on System Identification (SYSID)

作者： Sassella, Andrea Breschi, Valentina Formentin, Simone Politecn Milan Dipartimento Elettron Informaz & Bioingn Milan Italy Eindhoven Univ Technol Dept Elect Engn Eindhoven Netherlands

Retrieving optimal control actions in a receding horizon fashion at run time might be a challenging task, especially when the sampling time of the system to be controlled is small and the optimization problem is large. Although explicit solutions have been proposed to tackle this challenge, the complexity of the explicit control law scales poorly with the dimension of the problem. In the attempt to cope with these limitations within the challenging data-driven setup, we propose to construct a limited-complexity approximation of the explicit predictive law by iteratively exploring the state/reference space while leveraging structural priors on the input parameterization. The same approximation can be exploited to compute the control action also when the closed-loop system visits unexplored regions. The performance of the proposed strategy is assessed on a simple numerical example. Copyright (C) 2024 The Authors. This is an open access article under the CC BY-NC-ND license (https://***/licenses/by-nc-nd/4.0)

关键词： data-driven control

来源：评论

学校读者我要写书评

暂无评论

Finite Set Model Predictive control for PWM Rectifiers Based on data-driven Neural Network Predictor

Finite Set Model Predictive Control for PWM Rectifiers Based...

引用

ieee International Symposium on Circuits and systems (ISCAS)

作者： Liu, Lu Shi, Tao Wang, Dan Gu, Nan Peng, Zhouhua Dalian Maritime Univ Sch Marine Elect Engn Dalian 116026 Peoples R China Dalian Key Lab Swarm Control & Elect Technol Inte Dalian 116026 Peoples R China State Key Lab Maritime Technol & Safety Dalian Peoples R China

ISBN: (纸本)9798350330991;9798350331004

In this paper, a finite set model predictive control method based on data-driven neural network predictors (DNNPs) is proposed for pulse width modulation (PWM) rectifiers with fully unknown parameters. First, DNNPs are structured based on concurrent learning such that model uncertainties and input gains are identified simultaneously. Secondly, based on the information estimated by the predictors, a finite set model predictive power controller is designed, which is responsible for simplifying the rolling optimization and reducing the computational complexity. Finally, the stability analysis is provided based on input-to-state stability theory, and simulation results are provided to prove the effectiveness of the proposed method.

关键词： PWM Rectifier Finite set predictive control data-driven neural network predictor

来源：评论

学校读者我要写书评

暂无评论

A Method Based on Lie Group Machine learning for Multivariate Time-series Clustering 6

A Method Based on Lie Group Machine Learning for Multivariat...

引用

6th International conference on data-driven Optimization of Complex systems (DOCS)

作者： Huang, Yini Lu, Xiaopeng Guangdong Univ Technol Guangzhou Peoples R China Guangzhou Huali Coll Guangzhou Peoples R China

ISBN: (纸本)9798350377859;9798350377842

A multivariate time series (MTS) is a data series formed from observations of multiple variables at multiple time points, which may exhibit interdependencies and temporal dependencies. The high dimension and complex structure of MTS data present challenges to existing clustering methods in terms of feature dimension, computational complexity, and accuracy. To address these issues, we introduce a Lie Group machine learning method and propose a novel multivariate Timeseries Clustering method based on Lie Group Intrinsic Mean (T-CLGIM) to enhance clustering performance. We conducted extensive experiments on eight public and challenging MTS datasets. The results demonstrate that our method significantly outperforms state-of-the-art methods.

关键词： Multivariate Time Series Lie Group Manifold Intrinsic Mean Clustering

来源：评论

学校读者我要写书评

暂无评论

Improving Cooperation via Joint Intrinsic Motivation Exploration in Multi-Agent Reinforcement learning 6

Improving Cooperation via Joint Intrinsic Motivation Explora...

引用

6th International conference on data-driven Optimization of Complex systems (DOCS)

作者： Du, Hongfei Yang, Zhiyou Wu, Yunhan Xie, Ning Univ Elect Sci & Technol China Sch Comp Sci & Engn Chengdu 611731 Sichuan Peoples R China

ISBN: (纸本)9798350377859;9798350377842

Multi-agent reinforcement learning (MARL) encounters the enduring challenge of sparse rewards, which becomes particularly apparent in scenarios requiring coordinated actions among agents. To handle this issue, we consider adding an intrinsic reward to the environmental reward for enhancing the policy exploration capabilities in multi-agent cooperation settings. Our approach focuses on incentivizing strategic behaviors characterized by collective novelty among agents. Specifically, we introduce a self-supervised learning model to measure the novelty of diverse coordination patterns within a team of agents. To this end, we present a multi-agent intrinsic motivation framework called Joint Intrinsic Motivation Exploration (JIME) that adheres to the centralized learning with decentralized execution paradigm. Empirical evaluations demonstrate the crucial role of JIME in addressing tasks that require intricate coordination for optimal strategy execution. Our findings underscore the significance of incorporating intrinsic motivation mechanisms in MARL systems to facilitate effective collaboration among agents.

关键词： Multi-agent cooperation Reinforcement learning Intrinsic motivation

来源：评论

学校读者我要写书评

暂无评论

data-driven Dynamic Input Transfer for learning control in Multi-Agent systems with Heterogeneous Unknown Dynamics 62

Data-Driven Dynamic Input Transfer for Learning Control in M...

引用

62nd ieee conference on Decision and control (CDC)

作者： Lehmann, Dustin Drebinger, Philipp Seel, Thomas Raisch, Joerg Tech Univ Berlin Control Syst Grp Berlin Germany Tech Univ Berlin Res Cluster Excellence Sci Intelligence Berlin Germany FAU Erlangen Nurnberg Dept AIBE Intelligent Sensorimotor Syst Lab Erlangen Germany

ISBN: (纸本)9798350301243

learning input signals that make a dynamic system respond with a desired output is often data intensive and time consuming. It is therefore natural to ask whether, in a heterogeneous multi-agent scenario, an input signal learned by one agent can be suitably adapted and transferred to make the other agents respond with the same desired output, despite exhibiting different dynamics. In this paper, we propose a novel method to achieve this by employing a dynamic input transfer map. The method does not require any a-priori knowledge of the individual agents' dynamics. Instead, a small amount of experimental data from the source and target systems are used to estimate the transfer map. We evaluate the proposed method and compare it to existing approaches using static input transfer maps by investigating two example scenarios: (i) a simulation scenario for muscle dynamics, (ii) an experimental setting with a group of two-wheeled inverted pendulum robots and a sim-to-real motion learning problem.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Exponential Weighted Moving Average-Based Variational Mode Decomposition Method for Fault Diagnosis of Vehicle Batteries 13

Exponential Weighted Moving Average-Based Variational Mode D...

引用

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

作者： Wu, Penghao Tao, Hongfeng Tian, Engang Yin, Tangwen Chen, Hongtian Chen, Yiyang School of Mechanical and Electrical Engineering Soochow University Suzhou215137 China Jiangnan University Key Laboratory of Advanced Process Control for Light Industry of Ministry of Education Wuxi214122 China School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai200093 China Shanghai Jiao Tong University Department of Automation Shanghai200240 China

ISBN: (纸本)9798350361674

In recent years, the rapid advancement of electric vehicles has heightened concerns regarding the safety of high-energy batteries. Consequently, there has been a significant focus on the development of fault diagnosis technologies for batteries. Intelligent fault diagnosis (IFD), particularly data-driven approaches, represents a crucial methodology in addressing this concern. This paper proposes a data-driven approach, for IFD in high-energy batteries utilized in electric vehicles. The proposed Exponentially Weighted Average Variational Mode Decomposition (EWVMD) method overcomes these limitations by employing the exponential weighted moving average (EWMA) technique to smooth collected time-series voltage data. This process reduces noise, accentuates abnormal change trends, and yields secondary data with enhanced features. Subsequently, variational mode decomposition (VMD) is applied to the secondary data. The resulting decomposed sub-mode curves (intrinsic mode functions - IMFs) and the change trends of the residual curve, in conjunction with a fault threshold tailored to actual working conditions, are utilized to ascertain the occurrence of faults. Design verification experiments, the C42MSA electric vehicle power supply battery undergoes charging and discharging tests with injected faults at fixed intervals. The collected fault signals are then subjected to fault diagnosis using EWVMD. Results demonstrate the algorithm's superior detection effectiveness and increased sensitivity to fault signals compared to traditional methods. © 2024 ieee.

关键词： Variational mode decomposition

来源：评论

学校读者我要写书评

暂无评论

Analysis and Detectability of Offline data Poisoning Attacks on Linear Dynamical systems 5

Analysis and Detectability of Offline Data Poisoning Attacks...

引用

5th Annual conference on learning for Dynamics and control

作者： Russo, Alessio KTH Royal Inst Technol Div Decis & Control Syst Stockholm Sweden

In recent years, there has been a growing interest in the effects of data poisoning attacks on data-driven control methods. Poisoning attacks are well-known to the Machine learning community, which, however, make use of assumptions, such as cross-sample independence, that in general do not hold for linear dynamical systems. Consequently, these systems require different attack and detection methods than those developed for supervised learning problems in the i.i.d. setting. Since most data-driven control algorithms make use of the least-squares estimator, we study how poisoning impacts the least-squares estimate through the lens of statistical testing, and question in what way data poisoning attacks can be detected. We establish under which conditions the set of models compatible with the data includes the true model of the system, and we analyze different poisoning strategies for the attacker. On the basis of the arguments hereby presented, we propose a stealthy data poisoning attack on the least-squares estimator that can escape classical statistical tests, and conclude by showing the efficiency of the proposed attack. The code can be found here https://***/rssalessio/data- poisoning-linear-systems.

关键词： data poisoning data corruption data-driven control linear 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：