检索结果-内蒙古大学图书馆

17th International conference on Management of Large-Scale System Development, MLSD 2024

作者： Kulida, Elena Lebedev, Valentin V.A. Trapeznikov Institute of Control Sciences of Ras Laboratory of Decision Support Systems Moscow Russia

Approaches to detecting data anomalies using autoencoders and predicting the remaining useful life of aircraft engines using data-driven deep learning models are reviewed. The problem of planning aircraft engine maint... 详细信息

ISBN: (纸本)9798350375718

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Edge-based Parametric Digital Twins for Intelligent Building Indoor Climate Modeling 20

Edge-based Parametric Digital Twins for Intelligent Building...

引用

20th International conference on Factory Communication systems (WFCS)

作者： Ni, Zhongjun Zhang, Chi Karlsson, Magnus Gong, Shaofang Linkoping Univ Dept Sci & Technol Campus Norrkoping Linkoping Sweden Univ Gothenburg Dept Comp Sci & Engn Gothenburg Sweden

ISBN: (纸本)9798350319347;9798350319354

Digital transformation in the built environment generates vast data for developing data-driven models to optimize building operations. This study presents an integrated solution utilizing edge computing, digital twins, and deep learning to enhance the understanding of climate in buildings. Parametric digital twins, created using an ontology, ensure consistent data representation across diverse service systems equipped by different buildings. Based on created digital twins and collected data, deep learning methods are employed to develop predictive models for identifying patterns in indoor climate and providing insights. Both the parametric digital twin and deep learning models are deployed on edge for low latency and privacy compliance. As a demonstration, a case study was conducted in a historic building in Ostergotland, Sweden, to compare the performance of five deep learning architectures. The results indicate that the time-series dense encoder model exhibited strong competitiveness in performing multi-horizon forecasts of indoor temperature and relative humidity with low computational costs.

关键词： edge computing digital twin deep learning building indoor climate

来源：评论

学校读者我要写书评

暂无评论

A Velocity Tracking Method for Quadruped Robot with Rhythm controller 13

A Velocity Tracking Method for Quadruped Robot with Rhythm C...

引用

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

作者： Zhang, Xiaoping Wu, Yitong Xiu, Zhongchen School of Electrical and Control Engineering North China University of Technology Beijing100144 China University of Cambridge Department of Engineering CambridgeCB2 1PZ United Kingdom

ISBN: (纸本)9798350361674

In the motion performance of quadruped robots, the velocity is a very important item. It is hoped that a quadruped robot can move stably in accordance with the expected velocity. In this work, a velocity autonomous tracking method is proposed to realize real-time velocity tracking during the movement of quadruped robots. Firstly, the gait learning process of a quadruped robot is completed based on deep reinforcement learning, and the velocity adaptation ranges of different gaits are obtained. After that, the gait selection and velocity tracking controller selects the gait according to the desired velocity firstly, and makes periodic dynamic adjustment according to the velocity error signal after, thus completing the velocity tacking of the robot. The effectiveness of the method is verified by simulation experiments. The quadruped robot can well complete the velocity tracking task under different gaits, and the movement is relatively stable. © 2024 ieee.

关键词： Multipurpose robots

来源：评论

学校读者我要写书评

暂无评论

learning Switched Koopman Models for control of Entity-Based systems 62

Learning Switched Koopman Models for Control of Entity-Based...

引用

62nd ieee conference on Decision and control (CDC)

作者： Blischke, Madeline Hespanha, Joao P. Univ Calif Santa Barbara Santa Barbara CA 93106 USA

ISBN: (纸本)9798350301243

One problem that arises in control engineering is that of controlling a system for which the dynamics are unknown. Such a problem favors a data-driven approach, such as can be done through the use of the Koopman operator. We present a switched Koopman model, applicable to systems with discrete sets of inputs, that gives rise to an optimal control problem with a piecewise affine value function. This structure provides an efficient representation and enables a heuristic pruning algorithm that avoids the exponential complexity of finding the true optimal solution. We use density-like observables that are defined through the notion of entity-based systems: systems whose state is composed of a possibly varying number of entities that can be grouped into classes of like-entities. This encompasses many systems, including arcade games, a common benchmark used in reinforcement learning. We find that our approach requires much less training than commonly used for reinforcement learning. The Koopman approach also has the advantage of being agnostic to the control objective, which allows the objective to be changed without needing to retrain the model.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Novel PIML Architecture with Innovative learning Paradigm Applied in Battery Prognostics 10

A Novel PIML Architecture with Innovative Learning Paradigm ...

引用

10th International conference on control, Decision and Information Technologies (CoDIT)

作者： Deng, Weikun Nguyen, Khanh T. P. Gogu, Christian Mejaher, Kamal Morio, Jerome Le, Hung Wu, Dazhong Univ Technol Tarbes Occitanie Pyrenees Lab Genie Prod Tarbes France Univ Toulouse ISAE SUPAERO ICA Toulouse France Univ Toulouse ONERA DTIS Toulouse France Univ Cent Florida Addit Mfg & Intelligent Syst Lab Orlando FL 32816 USA

ISBN: (纸本)9798350373981;9798350373974

Prognostics and health management (PHM) increasingly play a constructive role throughout the entire lifetime of industrial equipment, significantly benefiting from extensive research in physical modelling and machine learning techniques. This has led to the development of hybrid approaches that seamlessly integrate both domains through physics-informed machine learning (PIML). PIML ensures the generation of cohesive solutions encompassing various aspects of physics knowledge across different stages of the machine-learning pipeline, substantially contributing to detection, diagnostics, and prognostics. However, PIML's design relies heavily on expert experience and demands rigorous interdisciplinary expertise, requiring a profound understanding of machine learning and physical principles. Inadequate design of PIML often leads to suboptimal outcomes, where the combined effect is less than the sum of its parts. Currently, PIML lacks a scalable and engineered application architecture to effectively utilise its embedded results. To address this challenge, our paper introduces a novel parallel architectural approach that employs pre-training and fine-tuning strategies for optimising the different model parts. Its data-driven branch is first trained with zero output in the PI branch, then fine-tuning the physics-informed branch. It takes the frozen data-driven model as a fixed feature extractor to get physics-consistency prediction. This approach proposes a generic solution for embedding physics knowledge into ML that guarantees performance improvement. The effectiveness of our approach is validated in the context of Remaining Useful Life (RUL) prediction using MIT-Stanford battery data.

关键词： Physics-informed machine learning architectural approach gainful learning strategy prognostic and health management pre-trained model physics consistency alignment fine-tuning

来源：评论

学校读者我要写书评

暂无评论

VDKD: A ViT-Based Student-Teacher Knowledge Distillation for Multi-Texture Class Anomaly Detection 13

VDKD: A ViT-Based Student-Teacher Knowledge Distillation for...

引用

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

作者： Chen, Yibo Peng, Haolong Zhu, Ke Zhang, Jianming College of Control Science and Engineering Zhejiang University Hangzhou310027 China Zhejiang Polytechnic Institute Polytechnic Institute Zhejiang University Hangzhou310027 China Zhejiang University Robotics Institute Yuyao314500 China

ISBN: (纸本)9798350361674

'With the progression of network perception and detection capabilities, multi-class anomaly detection and localization have become promising research areas. The primary challenge lies in mitigating minor textural defects while maintaining adaptability to a variable external environment. In response to this, we introduce a state-of-the-art pipeline, the ViT-based Discriminative Knowledge Distillation (VDKD). Contrary to existing lightweight models, VDKD uniquely leverages a robust teacher network to train a compact student model using innovative modules. Furthermore, the integrated Semantic Alignment Module (SAM) and Abnormal Assessment Module (AAM) assist in distilling essential information. We also propose the Discriminative Feature Judgment Module (DFJM), designed to emulate the teacher's output within the student model with limited sample intervention. Extensive self-supervised experiments on texture datasets, including KSDD2, MT, and MVTec-AD, underscore the effectiveness of our approach with VDKD. © 2024 ieee.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Improved YOLOv5 Algorithm for Special People Recognition in a New Guide Robot Dog for the Blind 13

Improved YOLOv5 Algorithm for Special People Recognition in ...

引用

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

作者： Liu, Shida Bi, Yu Ji, Honghai Wang, Li School of Electrical and Control Engineering North China University of Technology Beijing100000 China

ISBN: (纸本)9798350361674

Aiming at the problem of target recognition accuracy of the guide quadruped robot dog, this paper proposes a novel YOLO (SG-YOLOv5) detection and recognition algorithm based on changing the loss function and activation function. The proposed SG-YOLOv5 method replaces the activation function ReLU with SiLU and the loss function CIoU with GIoU in the YOLOv5 model. In addition, a database of special populations containing four categories, namely, blind family members, doctors, firefighters, and traffic police officers, is trained and applied to a real Unitree Go1 quadruped robot dog. The experimental results show that the recognition accuracy of the blind-guiding robot dog for special people reaches 94.3%, the check-all rate reaches 91.7%, and the average accuracy is as high as 95.1%, which verifies the effectiveness of the algorithm. © 2024 ieee.

关键词： Multipurpose robots

来源：评论

学校读者我要写书评

暂无评论

data-driven Based Model-Free Adaptive Optimal control Method for Hypersonic Morphing Vehicle

引用

ieee TRANSACTIONS ON AEROSPACE AND ELECTRONIC systems 2023年第4期59卷 3713-3725页

作者： Bao, Cunyu Wang, Peng Tang, Guojian Natl Univ Def Technol Coll Aerosp Sci & Engn Changsha 410073 Peoples R China

Aiming at the attitude control of a class of hypersonic morphing vehicles (HMVs) with variable sweep wings, a model-free adaptive dynamic planning (MFADP) optimal control method based on data-driven and finite-time fuzzy disturbance observer is proposed in this article. An integrated oriented-control attitude-morphing model is established, and morphing is considered as the state to carry out the integrated coupling attitude control. The control scheme is organized by a steady-feedback-compensation framework. To overcome the dependence on the unknown dynamic knowledge of HMVs, the dynamic model is reconstructed using neural networks for steady-state control. Subsequently, based on the MFADP algorithm, an Off-On serial policy learning strategy is designed for the error model to obtain a real-time approximate optimal feedback control. Additionally, a fuzzy disturbance observer with finite-time convergence ability is proposed to estimate and compensate the multiple uncertainties. Finally, the stability of the closed-loop system is proved theoretically and the simulation results demonstrate the improved performance of the proposed method.

关键词： Aerodynamics Adaptation models Optimal control Attitude control Steady-state Vehicle dynamics Uncertainty

来源：评论

学校读者我要写书评

暂无评论

Additive-state-decomposition-Based Event Triggered Spatial Repetitive control for Fast Tool Servo systems 13

Additive-state-decomposition-Based Event Triggered Spatial R...

引用

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

作者： Yang, Qin Zhou, Lan Xiao, Wenbin Li, Meiliu School of Information and Electrical Engineering Hunan University of Science and Technology Xiangtan411100 China

ISBN: (纸本)9798350361674

This paper presents an event triggered spatial repetitive controller (ETSRC) method for a fast tool servo (FTS) system with multiple disturbances to improve the tracking performance for position-dependent periodic signals, which is based on the additive state decomposition (ASD) technique and the disturbance observer (DOB) approach. First, the original tracking problem for a position-dependent periodic reference input of the disturbed FTS system is decomposed into two sub-problems, namely, the tracking problem of the linear time-invariant (LTI) primary system and the robust stabilization problem of the secondary system. For the LTI primary system, an ETSRC method is constructed to ensure the tracking capability of the system while saving the communication and computational resources. Secondly, in the secondary system, the available position information of the FTS system is used to design a DOB to estimate and compensate for multiple disturbances. Finally, comparative simulations are performed on the FTS system and the results show that the method outperforms the conventional DOB control method and other repetitive control methods. © 2024 ieee.

关键词： Disturbance rejection

来源：评论

学校读者我要写书评

暂无评论

Yolo-S: Texture-Less Object Pose Estimation with Shape Prior 13

Yolo-S: Texture-Less Object Pose Estimation with Shape Prior

引用

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

作者： Wang, Danzhu Liu, Shan Polytechnic Institute Zhejiang University Hangzhou310015 China College of Control Science and Engineering Zhejiang University State Key Laboratory of Industrial Control Technology Hangzhou310027 China

ISBN: (纸本)9798350361674

This paper aims to address the challenge of pose estimation for texture-less objects under conditions of stacking, occlusion, and complex backgrounds. The proposed method, called Yolo-S, consists of two parts: a pre-estimation network based on Yolov8 backbone and a key point post-refinement network based on Graph Neural Network. The first part uses off-the-shelf detectors result as inputs (e.g. MaskRCNN) and outputs the pre-estimate keypoints at the pixel coordinates. The second part encodes the roughly estimated keypoints and the priori knowledge of the object's CAD model, uses a graph neural network to intergrate feature and predict the final keypoints. PNP method is used to obtain the object's pose. The main advantage of Yolo-S is that is has a prior knowledge embedded GNN model for estimating the pose of texture-less objects, which greatly improves prediction quality and accuracy and achieves zero-shot sim-to-real model transfer. Experimental results on the LINEMOD dataset demonstrate the effectiveness of the proposed method and its significant competitiveness with other state-of-the-art methods. © 2024 ieee.

关键词： Graph neural networks

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：