检索结果-内蒙古大学图书馆

Black-box domain adaptation for cross-domain on-device machinery fault diagnosis via hierarchical debiased self-supervised learning

引用

Knowledge-Based systems 2025年 324卷

作者： Zhu, Mengliang Liu, Jie Lu, Yanglong Hu, Zhongxu Yuan, Li Zhou, Kaibo MOE Key Laboratory of Image Information Processing and Intelligent Control Belt and Road Joint Laboratory on Measurement and Control Technology School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan430074 China National Center of Technology Innovation for Digital Construction School of Civil and Hydraulic Engineering Huazhong University of Science and Technology Wuhan430074 China Department of Mechanical and Aerospace Engineering The Hong Kong University of Science and Technology Clear Water Bay Hong Kong SAR China School of Mechanical Science and Engineering Huazhong University of Science and Technology Wuhan430074 China

This paper explores black-box domain adaptation (BBDA) for cross-domain on-device machinery fault diagnosis. Specifically, a pre-trained black-box source model is deployed on a cloud platform with only its input-output API accessible, and a randomly initialized target model is trained locally utilizing unlabeled target data on an edge device. In BBDA, neither raw data nor model parameters are transmitted across domains, and the only available supervised information for adaptation is API-queried predictions on unlabeled target data. However, API-queried predictions are inevitably noisy and class-imbalanced due to cross-domain distribution discrepancy and inherent category bias. Directly utilizing them for on-device training easily leads to confirmation bias and exacerbates category bias, thereby severely degrading performance. This paper proposes a hierarchical debiased self-supervised learning framework for BBDA from a novel research perspective of concurrently alleviating both confirmation and category biases. Specifically, debiased knowledge distillation is proposed to gradually enhance discriminability of the randomly initialized target model by providing debiased training predictions. An adaptive category-unbiased sample division strategy is devised to divide unlabeled target data into certain-aware and uncertain-aware sets, enabling fully exploiting intrinsic data structures and utilizing underlying sample relationships within the target domain. On this basis, debiased prototypical self-training is proposed for the certain-aware set to obtain reliable pseudo-labels for self-training, leveraging global semantic structure. Hierarchical neighborhood adaptation is customized for the uncertain-aware set to enforce the cluster assumption for noise refinement, exploiting local clustering structure and utilizing underlying sample relationships. Extensive experiments and analyses on public and practical datasets validate the proposed framework's superiority in cross-dom

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

End-to-end Flow Scheduling Optimization for Industrial 5G and TSN Integrated Networks

End-to-end Flow Scheduling Optimization for Industrial 5G an...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Houling Liu Mingyan Li Fuqiang Gu Qihao Li Weiting Zhang Songtao Guo College of Computer Science Chongqing University P. R. China Key Laboratory of System Control and Information Processing Ministry of Education of China P. R. China College of Communication Engineering Jilin University Changchun P. R. China School of Electronic and Information Engineering Beijing Jiaotong University Beijing P. R. China

ISBN: (数字)9798350351255

ISBN: (纸本)9798350351262

The integrated of the fifth generation (5G) and time-sensitive networking (TSN) is a promising approach to meet the requirements of deterministic forwarding with extremely low latency and high flexibility for the Industrial Internet of Things (IIoT). However, due to the large dissimilarity of protocol stacks, the 5G system normally serves as a logical bridge for TSN, performing hold-and-forward operations in Base Stations (BSs) under the current 5G-TSN frameworks. To provide ubiquitous and seamless connectivity for IIoT devices, this paper focuses on the integrated enhancement of 5G and TSN by optimizing the scheduling of time-sensitive flows with end-to-end latency of 5G-TSN transmission taken into consideration. Specifically, we propose a novel architecture named GF-CQF, which combines 5G Grant-free (GF) access with TSN Cyclic Queuing and Forwarding (CQF). Subsequently, the distributed flow scheduling problem based on GF-CQF architecture is established due to the high timeliness demands. To alleviate the impact of the uncertainty inherent in 5G channels on end-to-end deterministic transmission, a feature-aware decentralized real-time scheduling (FDRS) policy based on Multi-agent Reinforcement Learning is proposed. FDRS allows each agent at BS to adaptively allocate TSN injection slots for flows mainly based on dynamic 5G transmission performance and TSN network resource state so that TSN queue overflow can be avoided and flow delay constraints can be guaranteed. Simulations show that FDRS offers superior scheduling capabilities under the limited TSN queue resources.

关键词： Wireless communication Training Job shop scheduling Uncertainty Protocols 5G mobile communication Reinforcement learning Real-time systems Resource management Industrial Internet of Things

来源：评论

学校读者我要写书评

暂无评论

Autoregressive GNN-ODE GRU Model for Network Dynamics

arXiv

引用

arXiv 2022年

作者： Liang, Bo Wang, Lin Wang, Xiaofan Department of Automation Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai200240 China Shanghai Key Laboratory of Power Station Automation Technology School of Mechatronic Engineering and Automation Shanghai University Shanghai China

Revealing the continuous dynamics on the networks is essential for understanding, predicting, and even controlling complex systems, but it is hard to learn and model the continuous network dynamics because of complex and unknown governing equations, high dimensions of complex systems, and unsatisfactory observations. Moreover, in real cases, observed time-series data are usually non-uniform and sparse, which also causes serious challenges. In this paper, we propose an Autoregressive GNN-ODE GRU Model (AGOG) to learn and capture the continuous network dynamics and realize predictions of node states at an arbitrary time in a data-driven manner. The GNN module is used to model complicated and nonlinear network dynamics. The hidden state of node states is specified by the ODE system, and the augmented ODE system is utilized to map the GNN into the continuous time domain. The hidden state is updated through GRUCell by observations. As prior knowledge, the true observations at the same timestamp are combined with the hidden states for the next prediction. We use the autoregressive model to make a one-step ahead prediction based on observation history. The prediction is achieved by solving an initial-value problem for ODE. To verify the performance of our model, we visualize the learned dynamics and test them in three tasks: interpolation reconstruction, extrapolation prediction, and regular sequences prediction. The results demonstrate that our model can capture the continuous dynamic process of complex systems accurately and make precise predictions of node states with minimal error. Our model can consistently outperform other baselines or achieve comparable performance. Copyright © 2022, The Authors. All rights reserved.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

3d Surface Detection and Segmentation from Point Clouds Via Quadric Fits Based on Dbscan Clustering

SSRN

引用

SSRN 2024年

作者： Xie, Tingting Chen, Hui Liu, Wanquan Zhou, Rongyu Li, Qilin College of Automation Engineering Shanghai University of Electric Power Shanghai200090 China Key Laboratory of System Control and Information Processing Ministry of Education Shanghai200240 China School of Intelligent Systems Engineering Sun Yat-sen University Guangzhou510275 China School of Electrical Engineering Computing and Mathematical Sciences Curtin University BentleyWA Australia

Extracting surfaces from 3D point clouds is significant in reconstructing and transforming these discrete points into their corresponding models. Scannedpoint clouds are often accompanied by noise, and the existing methods mainly rely on local feature similarities for surface extractions. Errors in estimatingthe feature information may lead to incorrect surface detection. In this paper we propose a surface extraction and boundary detection method based onclustering technique. The method can be described in three steps: In the first step, a normal correction is carried out using the information from theneighbourhood of points with sharp features. The second step is to cluster the points that meet the coplanar condition of the local quadric surface (LQS).In the third step, surface merger is performed by merging the local surfaces satisfying the merging conditions. Experimental validation is carried outto determine the effectiveness of the proposed method. The experimental results show improved surface extraction accuracy of the proposed methodin comparison to RANSAC, RG, LCCP, C2NO and HT methods. © 2024, The Authors. All rights reserved.

关键词： Merging

来源：评论

学校读者我要写书评

暂无评论

GOGICP: A Real-time Gaussian Octree-based GICP Method for Faster Point Cloud Registration

GOGICP: A Real-time Gaussian Octree-based GICP Method for Fa...

引用

IEEE International Conference on Real-time Computing and Robotics (RCAR)

作者： Zhitao Yu Wei Yuan Hengwang Zhao Hanyang Zhuang Ming Yang Department of Automation Shanghai Jiao Tong University Shanghai China Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai China Global Institute of Future Technology Shanghai Jiao Tong University Shanghai China University of Michigan-Shanghai Jiao Tong University Joint Institute Shanghai Jiao Tong University Shanghai China

ISBN: (数字)9798350372601

ISBN: (纸本)9798350372618

Point cloud registration algorithms are crucial for robot localization based on prior maps. Classical methods often struggle to achieve both high accuracy and real-time performance simultaneously, as these two goals can conflict. This limitation impedes the application of point cloud registration in high-speed scenarios, such as the LiDAR localization of fast-moving robots, where both precision and real-time capabilities are essential. This paper proposes a real-time Gaussian Octree-based Generalized Iterative Closest Point (GICP) method, which leverages the censoring properties of the Gaussian Octree to expedite the nearest neighbor search and incorporates an incremental calculation of the Gaussian distribution to hasten the tree-building process. Experimental results show that the proposed method can achieve a 61.8% improvement in speed over GICP while maintaining accuracy comparable to that of GICP. The average registration accuracy is up to 5 cm, with an angular error of less than 0.2°, and the average registration time for a single frame is 77 ms, which satisfies the demands of practical application scenarios.

关键词： Point cloud compression Location awareness Accuracy Laser radar Octrees Semantics Gaussian distribution

来源：评论

学校读者我要写书评

暂无评论

Size-invariance matters: rethinking metrics and losses for imbalanced multi-object salient object detection 24

Size-invariance matters: rethinking metrics and losses for i...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Feiran Li Qianqian Xu Shilong Bao Zhiyong Yang Runmin Cong Xiaochun Cao Qingming Huang Institute of Information Engineering Chinese Academy of Sciences Beijing China and School of Cyber Security University of Chinese Academy of Sciences Beijing China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China Institute of Information Science Beijing Jiaotong University Beijing China and School of Control Science and Engineering Shandong University Jinan China and Key Laboratory of Machine Intelligence and System Control Ministry of Education Jinan China School of Cyber Science and Tech. Shenzhen Campus Sun Yat-sen University School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China and Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China and Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing China

This paper explores the size-invariance of evaluation metrics in Salient Object Detection (SOD), especially when multiple targets of diverse sizes co-exist in the same image. We observe that current metrics are size-sensitive, where larger objects are focused, and smaller ones tend to be ignored. We argue that the evaluation should be size-invariant because bias based on size is unjustified without additional semantic information. In pursuit of this, we propose a generic approach that evaluates each salient object separately and then combines the results, effectively alleviating the imbalance. We further develop an optimization framework tailored to this goal, achieving considerable improvements in detecting objects of different sizes. Theoretically, we provide evidence supporting the validity of our new metrics and present the generalization analysis of SOD. Extensive experiments demonstrate the effectiveness of our method. The code is available at https://***/Ferry-Li/SI-SOD.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Submodularity-based False Data Injection Attack Scheme in Multi-agent Dynamical systems

arXiv

引用

arXiv 2022年

作者： Luo, Xiaoyu Zhao, Chengcheng Fang, Chongrong He, Jianping The Department of Automation Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai200240 China The State Key Laboratory of Industrial Control Technology Institute of Cyberspace Research Zhejiang University China The Department of Electrical and Computer Engineering University of Victoria BC Canada

Consensus in multi-agent dynamical systems is prone to be sabotaged by the adversary, which has attracted much attention due to its key role in broad applications. In this paper, we study a new false data injection (FDI) attack design problem, where the adversary with limited capability aims to select a subset of agents and manipulate their local multi-dimensional states to maximize the consensus convergence error. We first formulate the FDI attack design problem as a combinatorial optimization problem and prove it is NP-hard. Then, based on the submodularity optimization theory, we show the convergence error is a submodular function of the set of the compromised agents, which satisfies the property of diminishing marginal returns. In other words, the benefit of adding an extra agent to the compromised set decreases as that set becomes larger. With this property, we exploit the greedy scheme to find the optimal compromised agent set that can produce the maximum convergence error when adding one extra agent to that set each time. Thus, the FDI attack set selection algorithms are developed to obtain the near-optimal subset of the compromised agents. Furthermore, we derive the analytical suboptimality bounds and the worst-case running time under the proposed algorithms. Extensive simulation results are conducted to show the effectiveness of the proposed algorithm. Copyright © 2022, The Authors. All rights reserved.

关键词： Dynamical systems

来源：评论

学校读者我要写书评

暂无评论

A Real-time Framework of Multi-Camera Vehicle Tracking system

A Real-time Framework of Multi-Camera Vehicle Tracking Syste...

引用

2024 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2024

作者： Zhuang, Hanyang Chen, Xingyuan Lan, Jianglin Wang, Ying Lu, Zhaoxuan Wang, Chunxiang Yang, Ming University of Michigan Shanghai Jiao Tong University Joint Institute Shanghai Jiao Tong University Shanghai200240 China University of Glasgow James Watt School of Engineering GlasgowG12 8QQ United Kingdom Shanghai Jiao Tong University School of Mechanical Engineering Shanghai200040 China Shanghai Intelligent and Connected Vehicle R and D Center Co. LTD Shanghai201400 China Shanghai Jiao Tong University Department of Automation Shanghai200240 China Ministry of Education of China Key Laboratory of System Control and Information Processing Shanghai200240 China Shanghai Engineering Research Center of Intelligent Control and Management Shanghai200240 China

ISBN: (纸本)9798350372601

Multi-camera vehicle tracking (MCVT) system is a key technology to build an intelligent city and intelligent transportation system. The MCVT system utilizes roadside monitoring devices and computing platforms to achieve city-level road traffic perception at a finer scale by tracking each vehicle on the road to obtain the trajectory. The existing MCVT system mainly focuses on dealing with the algorithm modules separately, without studying the real-time performance and system architecture of the entire pipeline. However, in real-world applications, many of the tricks in improving evaluation metrics are not applicable since there is endless traffic flow running all the time. Therefore, this paper proposes a real-time framework of the MCVT system to streamline all the algorithm modules with proper data flow. Firstly, a multi-threaded multi-vehicle tracking (MVT) module is developed to detect vehicles in each camera. Re-identification is implemented during MVT to provide the feature for tracking in the same camera and across different cameras. Then, after the vehicle leaves the camera field of view, the feature of vehicle trajectory is calculated by averaging the feature at each frame. Meanwhile, a background thread runs continuously to match the incoming feature of vehicle trajectory with past features by conducting a topology searching method. In the end, the vehicle trajectories with well-matched features are considered to be the same vehicle. The entire pipeline has been validated on two different datasets, the precision is 0.64 in the CityFlow dataset which uses 6 cameras simultaneously. In the customized dataset, the same method can still reach real-time when using 16 cameras at the same time with smaller network modules. © 2024 IEEE.

关键词： Vehicle locating systems

来源：评论

学校读者我要写书评

暂无评论

A Retina-LGN-V1 Structure-like Spiking Neuron Network for Image Feature Extraction 21

A Retina-LGN-V1 Structure-like Spiking Neuron Network for Im...

引用

Proceedings of the 2021 5th International Conference on Video and Image processing

作者： Sheng Ji Qijun Gu Ye Yuan Peng Zhao Tao Fang Hong Huo Xin Niu Department of Automation_School of Electronic Information and Electrical Engineering_Key Laboratory of System Control and Information Processing Ministry of Education Shanghai Jiao Tong University China College of Computer Science and Technology on Parallel and Distributed Laboratory National University of Defense Technology China

ISBN: (纸本)9781450385893

SNNs have achieved great attention in recent years as they contain neurons more like those in the brain and use spikes to encode and transmit information efficiently among neurons with lower energy consumption. A Retina-LGN-V1 structure-like spiking neuron network (RLVSL-SNN) is proposed in this paper. It is inspired by the structure of mammalian primary visual pathway, and simulates different biological structures of Retina, LGN and V1. Noise reduction circuit and light adaptation circuit are also simulated for enhancing the robustness of its extracted features. RLVSL-SNN is a bio-plausible neuron network as it has firing rates of neurons in each layer that are similar to those of biological experiments. Besides, a full-connected SNN (FC SNN) is implemented following RLVSL-SNN for classification to evaluate the extracted features. The additive spiking timing dependent plasticity (STDP) learning rules and the ANN-to-SNN conversion method are utilized to train RLVSL-SNN and FC SNN, respectively. The experiments on MNIST dataset have verified that RLVSL-SNN is comparable to AlexNet for classification by features from spikes.

关键词： Additive Spiking Timing Dependent Plasticity Learning Primary Visual Pathway Image Classification Spiking Neural Network

来源：评论

学校读者我要写书评

暂无评论

Recurrent Attentional Reinforcement Learning for Machinery Fault Diagnosis

SSRN

引用

SSRN 2024年

作者： Tang, Zhenhui Wang, Jingcheng Wu, Shunyu The Department of Automation The Key Laboratory of System Control and Information Processing Ministry of Education of China The Shanghai Engineering Research Center of Intelligent Control and Management Shanghai Jiao Tong University No.800 Dongchuan Road Shanghai200240 China The SJTU Sanya Yazhou Bay Institute of Deepsea Science and Technology Sanya572024 China The Autonomous Systems and Intelligent Control International Joint Research Center Xi’an Technological University Xi’an710021 China

Recognizing fault types of machinery system is a fundamental but challenging task in industrial application. Although remarkable progress has been attained by learning fault features and predicting the corresponded fault type with deep neural networks, the existing fault diagnosis pipelines usually learn non-discriminative information. Contextual sequence dependency between the localized time fragments is often ignored. To deal with these problem, a recurrent attention framework with reinforcement learning is proposed to find a series of informative features of local time fragments related to different faults in an iterative manner. Fault labels are further predicted on these time fragments. The proposed framework also explicitly models long-term dependencies among these attentional time fragments to capture historical fault information. Experiments on two typical machinery systems, bearing system and hydraulic system, show that the proposed framework achieves superior diagnosis performance and increases no computation burden on CPU. © 2024, The Authors. All rights reserved.

关键词： Reinforcement learning

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：