检索结果-内蒙古大学图书馆

SS-ADA: A Semi-Supervised Active Domain Adaptation Framework for Semantic Segmentation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Yan, Weihao Qian, Yeqiang Li, Yueyuan Li, Tao Wang, Chunxiang Yang, Ming Department of Automation Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai200240 China

Semantic segmentation plays an important role in intelligent vehicles, providing pixel-level semantic information about the environment. However, the labeling budget is expensive and time-consuming when semantic segmentation model is applied to new driving scenarios. To reduce the costs, semi-supervised semantic segmentation methods have been proposed to leverage large quantities of unlabeled images. Despite this, their performance still falls short of the accuracy required for practical applications, which is typically achieved by supervised learning. A significant shortcoming is that they typically select unlabeled images for annotation randomly, neglecting the assessment of sample value for model training. In this paper, we propose a novel semi-supervised active domain adaptation (SS-ADA) framework for semantic segmentation that employs an image-level acquisition strategy. SS-ADA integrates active learning into semi-supervised semantic segmentation to achieve the accuracy of supervised learning with a limited amount of labeled data from the target domain. Additionally, we design an IoU-based class weighting strategy to alleviate the class imbalance problem using annotations from active learning. We conducted extensive experiments on synthetic-to-real and real-to-real domain adaptation settings. The results demonstrate the effectiveness of our method. SS-ADA can achieve or even surpass the accuracy of its supervised learning counterpart with only 25% of the target labeled data when using a real-time segmentation model. The code for SS-ADA is available at https://***/ywher/SS-ADA. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantic Segmentation

Chaos synchronization of semiconductor lasers over 1040-km fiber relay transmission with hybrid amplification

学校读者我要写书评

暂无评论

Photonics Research 2023年第6期11卷 953-960页

作者： LONGSHENG WANG JUNLI WANG YUSHAN WU YUEHUI SUN SONGSUI LI LIANSHAN YAN YUNCAI WANG ANBANG WANG Key Laboratory of Advanced Transducers and Intelligent Control System Ministry of EducationTaiyuan 030024China College of Physics and Optoelectronics Taiyuan University of TechnologyTaiyuan 030024China School of Information Engineering Guangdong University of TechnologyGuangzhou 510006China Guangdong Provincial Key Laboratory of Photonics Information Technology Guangzhou 510006China Center for Information Photonics&Communications School of Information Science and TechnologySouthwest Jiaotong UniversityChengdu 610031China

Optical chaos communication and key distribution have been extensively demonstrated with high-speed advantage but only within the metropolitan-area network range of which the transmission distance is restricted to around 300 *** secure-transmission requirement of the backbone fiber link,the critical threshold is to realize long-reach chaos ***,we propose and demonstrate a scheme of long-reach chaos synchronization using fiber relay transmission with hybrid amplification of an erbium-doped fiber amplifier(EDFA)and a distributed fiber Raman amplifier(DFRA).Experiments and simulations show that the hybrid amplification extends the chaos-fidelity transmission distance thanks to that the low-noise DFRA suppresses the amplified spontaneous emission noise and self-phase *** of the hybrid-relay conditions are studied,including launching power,gain ratio of DFRA to EDFA,single-span fiber length,and number of fiber span.A 1040-km chaos synchronization with a synchronization coefficient beyond 0.90 is experimentally achieved,which underlies the backbone network-oriented optical chaos communication and key distribution.

关键词： synchronization fiber amplifier

Co-Scheduling of Energy and Production in Discrete Manufacturing Considering Decision-Dependent Uncertainties

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Pan, Yiyuan Wang, Zhaojian The Key Laboratory of System Control and Information Processing Ministry of Education of China Department of Automation Shanghai Jiao Tong University Shanghai200240 China

Modern discrete manufacturing requires real-time energy and production co-scheduling to reduce business costs. In discrete manufacturing, production lines and equipment are complex and numerous, which introduces significant uncertainty during the production process. Among these uncertainties, decision-dependent uncertainties (DDUs) pose additional challenges in finding optimal production strategies, as the signature or the shape of the uncertainty set cannot be determined before solving the model. However, existing research does not account for decision-dependent uncertainties (DDUs) present in discrete manufacturing;moreover, current algorithms for solving models with DDUs suffer from high computational complexity, making them unsuitable for the real-time control requirements of modern industry. To this end, we proposed an energy-production co-scheduling model for discrete manufacturing, taking into account decision-dependent uncertainties (DDUs). Subsequently, we devised a method for transforming the constraints associated with DDUs into linear form ones. Finally, we designed a novel algorithm based on the column-and-constraint generation (C&CG) algorithm and undertook a theoretical analysis of its performance of convergence and algorithmic complexity. A real-world engine assembly line was used to test our model and algorithm. Simulation results demonstrate that our method significantly reduces production costs and achieves better frequency regulation performance. © 2024, CC BY.

关键词： Costs

Spatial-Temporal-Rule-based Traffic Movement Status Interpretation at Intersections *

学校读者我要写书评

暂无评论

Spatial-Temporal-Rule-based Traffic Movement Status Interpre...

2023 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2023

作者： Zhuang, Hanyang Gao, Liyang Qian, Yeqiang Wang, Chunxiang Yang, Ming University of Michigan Shanghai Jiao Tong University Joint Institute Shanghai Jiao Tong University 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)9798350327182

Traffic light is the most significant component of road transportation. Traffic lights control the traffic movement at intersections by allowing the movement of some lanes and forbidding that of the others to ensure safety and efficiency. It is important for intelligent vehicles to understand the meaning of traffic lights. Conventional research works on traffic light detection and tracking consider each light independently, so the occlusion of some lights may mislead the decision of the vehicle. Since all traffic lights at an intersection work together, it is more essential to understand the traffic movement status controlled by the traffic lights rather than the perception of a single light. Therefore, this paper proposes a spatial-temporal-rule-based framework to interpret the current traffic movement status and predict the next status. The spatial information, including position and class, is obtained by traffic light detection. The temporal information is perceived by the proposed topologic-feature-based tracking method. The rule information is recognized by combination of spatial and temporal information. A dual-view camera system consisting of a telephoto camera and a wide-angle camera is adopted to ensure long perception distance and wide coverage simultaneously. The successful tracking of traffic lights from 200 m to 10 m away provides enough data to recognize the phase-change rule, which leads to correct interpretation and prediction. A customized dataset was collected, and experiments were conducted to illustrate the capability of the proposed method. © 2023 IEEE.

关键词： Cameras

Health Condition Monitoring of Satellite Momentum Wheel Bearing Based on Canonical Variable Analysis and Sliding Interval Variance

学校读者我要写书评

暂无评论

Health Condition Monitoring of Satellite Momentum Wheel Bear...

Industrial Cyber-Physical systems (ICPS)

作者： Sirui Du Shumei Zhang Yang Zhao Tianjin Key Laboratory of Process Measurement and Control School of Electrical and Information Engineering Tianjin University Tianjin China Institute of Spacecraft System Engineering China Academy of Space Technology Beijing China

With the rapid development of space technology, the demand for satellite reliability is getting higher and higher. Momentum wheel is a key component of satellite attitude control system, and its reliability is an important factor affecting the life of satellite. The condition monitoring of momentum wheel bearing (MWB) is of great significance to ensure the long life and high reliable operation of the satellite. In this paper, a new monitoring method based on multivariate statistics and canonical variable analysis (CVA) is proposed, and a new health degree function is defined from both dynamic and static aspects. First, the time-frequency domain analysis technique is used to extract the features of MWB in time domain, frequency domain and time-frequency domain, and the multi-domain high-dimensional health condition feature set is constructed. Then, in order to reduce the complexity of the problem, feature reduction is realized based on CVA. On the basis of considering steady-state error and sliding interval variance (SIV), the health degree (HD) characterizing the performance condition of MWB is defined. Finally, the experimental results based on the bearing test-bed show that the proposed method is feasible and effective, and the data-driven health condition monitoring of satellite momentum wheel bearing is realized.

关键词： Condition monitoring Time-frequency analysis Analytical models Satellites Wheels Feature extraction Complexity theory

OIDM: An Observability-based Intelligent Distributed Edge Sensing Method for Industrial Cyber-Physical systems

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wang, Shigeng Jin, Tiankai Ma, Yehan Chen, Cailian Department of Automation Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai200240 China

Industrial cyber-physical systems (ICPS) integrate physical processes with computational and communication technologies in industrial settings. With the support of edge computing technology, it is feasible to schedule large-scale sensors for efficient distributed sensing. In the sensing process, observability is the key to obtaining complete system states, and stochastic scheduling is more suitable considering uncertain factors in wireless communication. However, existing works have limited research on observability in stochastic scheduling. Targeting this issue, we propose an observability-based intelligent distributed edge sensing method (OIDM). Deep reinforcement learning (DRL) methods are adopted to optimize sensing accuracy and power efficiency. Based on the system’s ability to achieve observability, we establish a bridge between observability and the number of successful sensor transmissions. Novel linear approximations of observability criteria are provided, and probabilistic bounds on observability are derived. Furthermore, these bounds guide the design of action space to achieve a probabilistic observability guarantee in stochastic scheduling. Finally, our proposed method is applied to the estimation of slab temperature in industrial hot rolling process, and simulation results validate its effectiveness. Copyright © 2024, The Authors. All rights reserved.

关键词： Stochastic systems

Predicting influenza-like illness trends based on sentinel surveillance data in China from 2011 to 2019:A modelling and comparative study

学校读者我要写书评

暂无评论

Infectious Disease Modelling 2024年第3期9卷 816-827页

作者： Xingxing Zhang Liuyang Yang Teng Chen Qing Wang Jin Yang Ting Zhang Jiao Yang Hongqing Zhao Shengjie Lai Luzhao Feng Weizhong Yang School of Population Medicine and Public Health Chinese Academy of Medical Sciences&Peking Union Medical CollegeBeijing100073China State Key Laboratory of Respiratory Health and Multimorbidity China Key Laboratory of Pathogen Infection Prevention and Control(Peking Union Medical College) Ministry of EducationChina National Key Laboratory of Intelligent Tracking and Forecasting for Infectious Diseases National Institute for Communicable Disease Control and PreventionChinese Center for Disease Control and PreventionChina Department of Management Science and Information System Faculty of Management and EconomicsKunming University of Science and TechnologyKunming650506China Department of Applied Mathematics&Statistics Stony Brook UniversityStony BrookNY11794-3600USA WorldPop School of Geography and Environmental ScienceUniversity of SouthamptonSouthamptonSO171BJUK

Background Influenza is an acute respiratory infectious disease with a significant global disease ***,the coronavirus disease 2019 pandemic and its related non-pharmaceutical interventions(NPIs)have introduced uncertainty to the spread of ***,comparative studies on the performance of innovative models and approaches used for influenza prediction are ***,this study aimed to predict the trend of influenza-like illness(ILI)in settings with diverse climate characteristics in China based on sentinel surveillance data using three approaches and evaluate and compare their predictive *** The generalized additive model(GAM),deep learning hybrid model based on Gate Recurrent Unit(GRU),and autoregressive moving average-generalized autoregressive conditional heteroscedasticity(ARMA—GARCH)model were established to predict the trends of ILI 1-,2-,3-,and 4-week-ahead in Beijing,Tianjin,Shanxi,Hubei,Chongqing,Guangdong,Hainan,and the Hong Kong Special Administrative Region in China,based on sentinel surveillance data from 2011 to *** relevant metrics,namely,Mean Absolute Percentage Error(MAPE),Root Mean Squared Error(RMSE),and R squared,were calculated to evaluate and compare the goodness of fit and robustness of the three *** Considering the MAPE,RMSE,and R squared values,the ARMA—GARCH model performed best,while the GRU-based deep learning hybrid model exhibited moderate performance and GAM made predictions with the least accuracy in the eight settings in ***,the models’predictive performance declined as the weeks ahead ***,blocked cross-validation indicated that all models were robust to changes in data and had low risks of *** Our study suggested that the ARMA—GARCH model exhibited the best accuracy in predicting ILI trends in China compared to the GAM and GRU-based deep learning hybrid ***,in the future,the ARMA—GARCH model may be used

关键词： Influenza-like illness Influenza Sentinel surveillance China Predicting Modeling

AMFD: Distillation via Adaptive Multimodal Fusion for Multispectral Pedestrian Detection

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Zizhao Qian, Yeqiang Yang, Xiaoxiao Wang, Chunxiang Yang, Ming Department of Automation Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai200240 China

Multispectral pedestrian detection has been shown to be effective in improving performance within complex illumination scenarios. However, prevalent double-stream networks in multispectral detection employ two separate feature extraction branches for multi-modal data, leading to nearly double the inference time compared to single-stream networks utilizing only one feature extraction branch. This increased inference time has hindered the widespread employment of multispectral pedestrian detection in embedded devices for autonomous systems. To efficiently compress multispectral object detection networks, we propose a novel distillation method, the Adaptive Modal Fusion Distillation (AMFD) framework. Unlike traditional distillation methods, the AMFD framework fully leverages the original modal features from the teacher network, thereby significantly enhancing the performance of the student network. Specifically, a Modal Extraction Alignment (MEA) module is utilized to derive learning weights for student networks, integrating focal and global attention mechanisms. This methodology enables the student network to acquire optimal fusion strategies independent from that of teacher network without necessitating an additional feature fusion module. Furthermore, we present the SMOD dataset, a well-aligned challenging multispectral dataset for detection. Extensive experiments on the challenging KAIST, LLVIP and SMOD datasets are conducted to validate the effectiveness of AMFD. The results demonstrate that our method outperforms existing state-of-the-art methods in both reducing log-average Miss Rate and improving mean Average Precision. The code is available at https://***/bigD233/***. Copyright © 2024, The Authors. All rights reserved.

关键词： Feature extraction

Resilient control of Networked control systems With Hidden DoS Attacks and Unknown Observation Probability

学校读者我要写书评

暂无评论

IEEE Transactions on control of Network systems 2025年第2期12卷 1313-1323页

作者： Wang, Yuan Yan, Huaicheng Park, Ju H. Zhang, Hao Shen, Hao East China University of Science and Technology School of Information Science and Engineering Shanghai200237 China Suzhou University of Science and Technology School of Electronic and Information Engineering Suzhou215009 China East China University of Science and Technology Key Laboratory of Smart Manufacturing in Energy Chemical Process of Ministry of Education Shanghai200237 China Weifang University College of Machinery and Automation Weifang261061 China Yeungnam University Department of Electrical Engineering Gyeongsan38541 Korea Republic of Tongji University Department of Control Science and Engineering Shanghai200092 China Anhui University of Technology School of Electrical and Information Engineering Ma'anshan243002 China

This paper systematically studies the problem of resilient controller design for networked control systems under imperfect communication environments. The denial of service (DoS) attacks in the case of an imperfect network are remodeled. The semi-Markov chain is introduced in this paper to provide a suitable description of random attacks. On this basis, the double-layer stochastic process is constructed, which consists of an attack sequence and an observation one. Second, since the attack modes are inaccessible, a controller is constructed based on the observation sequence by virtue of the emission probability to stabilize the system under random attacks. Due to the complexity of the network, the probability that an observer can accurately detect an attack when it occurs is considered unknown. Based on the introduced double-layer stochastic process, a mode-associated Lyapunov function is constructed to facilitate the stability analysis. Building on this foundation, sufficient conditions that can ensure the δ-error mean square stability of the closed-loop system are derived. In the end, a simulation of the unmanned ground vehicle is presented, showcasing the feasibility and validity of the proposed control scheme. © 2025 IEEE.

关键词： Stochastic systems