检索结果-内蒙古大学图书馆

G-Anomaly: A Pyramid Graph Transformer-based Vision-Language Model for General industrial Anomaly Detection

IEEE Transactions on Automation Science and engineering 2025年

作者： Li, Jiaqi Wen, Shuhuan Fang, Bin Yanshan University Key Lab of Industrial Computer Control Engineering of Hebei Province Qinhuangdao066004 China Beiing University of Posts and Telecommunications Department of Artificial intelligence Beijing China

Visual-language alignment is crucial for enhancing the domain adaptability of industrial anomaly detection models. However, the existing methods overlook the importance of structured image representation, fail to further distinguish topological differences between anomalies and the inherent textures of products, which reduces the accuracy of semantic matching. To address this problem, we propose a novel industrial anomaly detection model G-Anomaly, to preserve the topological structure of the sample images and further enhance the model’s domain adaptability. We designed Pyramid Graph Transformer as a visual encoder to extract multi-scale visual features, which can directly preserve the structural relationships between different regions of the image, and also optimize the over-smoothing issue present in deep graph networks, thereby retaining the distinguishability of anomalous nodes. Additionally, we design a Multi-level Domain Adapter that ensures semantic consistency of anomalous features across different scales and contexts by performing visual-language matching at various resolutions and levels of abstraction. This enhances the model’s domain adaptability for anomaly detection for a wide range of industrial products. We collect and craft an actual solar panel dataset PV_actual AD, and conduct extensive experiments on the public dataset MVTec AD as well as the actual solar panel dataset PV_actual AD. This has demonstrated that G-Anomaly not only performs well in standard testing environments but also exhibits robustness and domain adaptability for anomaly detection tasks in real-world scenarios. © 2025 IEEE. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Global-Local Context Enhanced Yolo for Small Object Detection in Uav Images

SSRN

引用

SSRN 2025年

作者： Wei, Lixin Han, Kangshuai Li, Xin Che, Haijun Hu, Ziyu Engineering Research Center of the Ministry of Education for Intelligent Control System and Intelligent Equipment Yanshan University Hebei Qinhuangdao066000 China Key Lab of Industrial Computer Control Engineering of Hebei Province Yanshan University Hebei Qinhuangdao066000 China

Object detection using Unmanned Aerial Vehicles (UAVs) has emerged as a criti cal application across diverse domains. However, the wide-angle views of drones often result in images containing a high density of small objects, posing chal lenges for object detection such as few learnable features, significant occlusion, and an imbalanced distribution of positive and negative samples. To address these issues, this paper introduces AGLC-YOLO, an enhanced version of the YOLOv7 architecture specifically designed for detecting small objects in UAV images. AGLC-YOLO integrates global and local context information through a Attention guide Global-Local Context Information Extraction (AGLC) module. This module employs parallel dilated convolutions to capture local context infor mation and a transformer-based structure to extract global dependencies, which are then fused using an improved attention mechanism. The network also adds an additional small object detection head to enrich the small object information in the model. Additionally, AGLC-YOLO utilizes an auxiliary bounding box in con junction with the Inner-Wise Intersection over Union (Inner-WIoU) loss function to accelerate the bounding box regression process and improve detection accu racy. Experimental results on the VisDrone and ManipalUav datasets demonstrate that AGLC-YOLO achieves significant improvements over the baseline YOLOv7 model, with anincrease of 3%inAP50and2.7%inAP95ontheVisDronedataset, and 1.9% in AP50 and 2% in AP95 on the ManipalUav dataset. © 2025, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Hierarchical Energy Management Strategy of Hybrid Electric Truck Fleet under Connected Environment

Hierarchical Energy Management Strategy of Hybrid Electric T...

引用

第43届中国控制会议

作者： Qizhe Lu Wenbin Tang Zitong Pang Yahui Zhang Xiaohong Jiao School of Electrical Engineering Yanshan University Key Lab of Industrial Computer Control Engineering of Hebei Province Yanshan University School of Mechanical Engineering Yanshan University

ISBN: (数字)9789887581581

ISBN: (纸本)9798350366907

Considering the slow-moving and energy-consuming issues that tend to occur when truck fleets drive,this paper proposes a hierarchical energy management strategy(EMS) for the hybrid electrical truck(HET) fleet,including speed planning and energy *** speed threshold range that allows the whole fleet to pass through the traffic light is computed in the upper-layer speed *** the model predictive control framework,a multi-objective optimization function is constructed for speed planning to find the optimum *** the lower-layer energy management,an offline model is established for power allocation between the Integrated Starter Generator and the Electric Motor and gear shift selection of automated manual transmission to reduce the dimensionality of the control ***,an adaptive equivalent consumption minimum strategy(A-ECMS) is designed based on the fleet's optimal speed *** the offline part,the transformation relationship between the co-state and equivalence factor is established based on Pontryagin's minimum principle(PMP),and PMP obtains the globally optimal control sequence in the historical driving cycle to train a back-propagation neural network(BPNN).The BPNN model predicts the current optimal co-state in the online part,transformed into a real-time optimal equivalence factor to guide the A-ECMS online *** simulation results indicate that the fuel economy of the proposed EMS is comparable to that of the global optimization method PMP and superior to that of Fixed EF-ECMS and ***,it enables the HET fleet to pass through traffic lights without stopping,thereby ensuring queue continuity and improving the passage efficiency of the fleet.

关键词： hybrid electric trucks equivalent consumption minimization strategy fleet speed planning back-propagation neural network model predictive control

来源：评论

学校读者我要写书评

暂无评论

Event-Triggered Robust Model Predictive control for Multi-Agent Systems Under Hybrid Attacks

Event-Triggered Robust Model Predictive Control for Multi-Ag...

引用

第42届中国控制会议

作者： Yongxin Yin Jianxiong Li Chongyi Gao Yiming Fang Key Lab of Industrial Computer Control Engineering of Hebei Province Yanshan University Department of Electromechanical Engineering Tangshan University

In this paper,the consensus control problem for multi-agent systems（MASs） suffering hybrid attacks is discussed,in which the hybrid attacks include randomly occurring deception attacks on system dynamic behavior and denial-of-service attacks（DoS）.An event-triggered robust model predictive control（RMPC） method is presented to achieve the consensus of MASs under hybrid *** can be proved that the consensus errors can be asymptotic with prescribed H performance,the sufficient condition for stability can be given in terms of linear matrix inequalities（LMIs）,and the controller gains are obtained by solving an optimization problem involving some *** the event-triggered mechanism with maximum trigger interval,not only can not only save computation time and facilitate the efficiency of the shared resources but also can effectively be responsible for the negative impact of cyber-attacks on system ***,simulation results are given to illustrate the validity and 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

A Novel A-ESO Composited Non-recursive control Methodology for Complex High-order Disturbed Nonlinear Systems

A Novel A-ESO Composited Non-recursive Control Methodology f...

引用

第42届中国控制会议

作者： Cong Zhang Shiji Liu Shuang Liu School of Electrical Engineering Yanshan University Key Lab of Industrial Computer Control Engineering of Hebei Province Yanshan University

This paper mainly focuses on a disturbance rejection control methodology for a series of mismatched and matched disturbed complex ***,a novel finite-time arbitrary-order extended state observer(A-ESO) is proposed to accurately estimate the system states,disturbances,and their corresponding order *** then,a novel concise non-recursive control ideology utilizing all the estimations of A-ESO is proposed to strictly guarantee finite-time stability and convergence of the whole closed-loop ***,compared with the existing non-recursive control methods,the essential characteristic is that all the estimation information is utilized to construct the composite *** will be tremendously conducive to cost-saving and promoting practical implementations.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Fault Tolerant control for the Vibration Displacement System of a Continuous Casting Mold Based on a Fault Detection Observer

Fault Tolerant Control for the Vibration Displacement System...

引用

第43届中国控制会议

作者： Wenqi Hao Yiming Fang Le Liu Xiaonan Liu Key Lab of Industrial Computer Control Engineering of Hebei Province Yanshan University Engineering Research Center of the Education Ministry for Intelligent Control System and Intelligent Equipment Yanshan University

ISBN: (数字)9789887581581

ISBN: (纸本)9798350366907

In this paper,a fault detection observer and an adaptive sliding mode fault-tolerant controller are designed for the vibration displacement system of a continuous casting mold with actuator faults and load torque ***,a mathematical model of a continuous casting mold vibration displacement system driven by a servo motor with actuator faults and load disturbance uncertainties are ***,a fault detection observer is designed to obtain information on actuator faults,load disturbances,and system state ***,based on the information from the fault detection observer,an adaptive sliding mode fault-tolerant controller is designed,and the stability of the designed system is proved using Lyapunov stability ***,the effectiveness of the designed adaptive sliding mode fault-tolerant controller based on fault observer is verified through simulation.

关键词： Continuous casting mold Fault detection observer Fault tolerant control Actuator fault Vibration displacement system Driven by a servo motor

来源：评论

学校读者我要写书评

暂无评论

Branch-Yolo：Efficient Object Detector for Thin Structure Objects Like Pantograph

SSRN

引用

SSRN 2024年

作者： Li, Yaqian Han, Jiaqi Li, Haibin Zhang, Wenming College of Electrical Engineering Yanshan University Hebei Qinhuangdao066000 China Key Laboratory of Industrial Computer Control Engineering of Hebei Province Hebei Qinhuangdao066000 China

The pantograph is the key structure in the railway Pantograph-OCS system, acting as the switch for the railway traction power supply system. So it’s crucial to detect the pantograph's lifting and lowing states for production efficiently and safely. However, due to the pantograph’s special slender structure and complex application scenarios, it occupies few pixels and is susceptible to background interference, making accurate real-time detection a challenging task. In this work, we aim to propose a generalized model, designed for detecting, branch-like objects with thin structure and multi-scale size such as pantographs. To this end, we propose the Branch-YOLO based on YOLOv8, which is improved in two stages, feature extraction and feature fusion, to simultaneously enhance perception. First, we introduce BrLayer is consist of BranchConv and EFAC(Extend receptive field attention Convolution), BranchConv can adaptively capture features of thin and tortuous local structures,EFAC is able to assist in expanding the effective receptive field. Subsequently, we propose a feature fusion network (BranchNet) to integrate multilevel semantic information and multi-scale features, which significantly reduces background interference in detecting slender structure objects and enhances the ability to perceive changes in object scale. Besides, we propose the Lightweight Cross-Kernel Convolution layer (LckLayer) and introduce the Feature Decomposition module (FdM) in BranchNet for lightweight design, reducing the computational overhead and enhancing model efficiency. Our proposed Branch-YOLO not only achieves the best performance on the multi-scale pantograph datasets but also attains an outstanding 42.9% AP on the COCO val2017 datasets, with 5.9M parameters and 16.8 GFLOPs. © 2024, The Authors. All rights reserved.

关键词： Religious buildings

来源：评论

学校读者我要写书评

暂无评论

CPL-SLAM: Centralized Collaborative Multi-Robot Visual-Inertial SLAM Using Point-and-Line Features

引用

IEEE Internet of Things Journal 2025年

作者： Liu, Xin Wen, Shuhuan Liu, Huaping Richard Yu, F. Yanshan University Engineering Research Center The Ministry of Education for Intelligent Control System and Intelligent Equipment Key Lab of Intelligent Rehabilitation and Neuroregulation in Hebei Province Department of Key Lab of Industrial Computer Control Engineering of Hebei Province Qinhuangdao066004 China Tsinghua University Department of Computer Science and Technology Beijing100084 China Shenzhen University College of Computer Science and Software Engineering China Carleton University Canada

Traditional visual-inertial SLAM (Simultaneous Localization and Mapping) systems predominantly rely on feature point matching from a single robot to realize the robot pose estimation and environment map construction. However, in complex scenarios, these traditional systems struggle with issues such as tracking failures due to illumination changes, rapid movements, and low-texture environments, and they perform poorly in terms of mapping efficiency and global consistency. To address these challenges, we propose a centralized collaborative SLAM system that employs both point and line features for tracking in the robot and map fusion in the cloud. The proposed system leverages the fusion of point and line features across all instances in the process, which allows our method to achieve higher localization accuracy in structured, low-texture scenes. With the aid of classifying gravity-aligned vertical lines and spatial parallel lines, the proposed system can deliver faster and more accurate odometry in complex scenes. Furthermore, we developed intra-robot and inter-robot loop closure detection methods based on point and line features, generating a globally consistent sparse point cloud and structured scene map in the cloud. Our method is able to build richer maps while improving accuracy compared to existing methods. Experimental results on public datasets and in real-world environments show that, compared to existing advanced methods, our approach demonstrates better performance. © 2014 IEEE.

关键词： SLAM robotics

来源：评论

学校读者我要写书评

暂无评论

A Novel A-ESO Composited Non-Recursive control Methodology for Complex High-Order Disturbed Nonlinear Systems

A Novel A-ESO Composited Non-Recursive Control Methodology f...

引用

Chinese control Conference (CCC)

作者： Cong Zhang Shiji Liu Shuang Liu School of Electrical Engineering Yanshan University Qinhuangdao Hebei China Key Lab of Industrial Computer Control Engineering of Hebei Province Yanshan University Qinhuangdao Hebei China

This paper mainly focuses on a disturbance rejection control methodology for a series of mismatched and matched disturbed complex systems. Firstly, a novel finite-time arbitrary-order extended state observer (A-ESO) is proposed to accurately estimate the system states, disturbances, and their corresponding order derivatives. And then, a novel concise nonrecursive control ideology utilizing all the estimations of A-ESO is proposed to strictly guarantee finite-time stability and convergence of the whole closed-loop system. Significantly, compared with the existing nonrecursive control methods, the essential characteristic is that all the estimation information is utilized to construct the composite controller. This will be tremendously conducive to cost-saving and promoting practical implementations.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A TCN-Transformer Model of Acceleration Prediction for Preceding Vehicle in CACC Systems

A TCN-Transformer Model of Acceleration Prediction for Prece...

引用

Conference on Vehicle control and Intelligence (CVCI)

作者： Wenzhao Tang Wenbin Tang Xiaohong Jiao Yahui Zhang Dongxue Zhang Hebei Key Laboratory of Industrial Computer Control Engineering Yanshan University Qinhuangdao Hebei Province China School of Mechanical Engineering Yanshan University Qinhuangdao China

ISBN: (数字)9798331504892

ISBN: (纸本)9798331504908

This study establishes an acceleration prediction model of the preceding vehicle tailored for cooperative adaptive cruise control (CACC) systems that address the common challenges posed by V2V communication delays or no communication from the preceding vehicle. The designed model integrates a temporal convolutional network (TCN) with the Transformer architecture block to create a sophisticated mechanism for temporal data analysis. This integration leverages the strengths of TCN in capturing spatial features and the multi-attention mechanism of Transformers to enhance the recognition of long-term dependencies in the time series of data. The synergy between these techniques not only effectively narrows the identification gap between short-term and long-term dependencies but also helps to process spatial information for accurate time-series prediction, significantly improving prediction accuracy. Simulations using the NGSIM dataset confirm the model's prediction performance over the traditional model. Further, the potential of the method to solve the communication delay problem is verified, highlighting the effectiveness of TCN and Transformer multi-attention mechanism in improving the CACC capability.

关键词： Adaptation models Accuracy Uncertainty Adaptive systems Time series analysis Vehicular ad hoc networks Predictive models Transformers Delays Cruise control

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：