检索结果-内蒙古大学图书馆

Toward Coordination Control of Multiple Fish-Like Robots:Real-Time Vision-Based Pose Estimation and Tracking via Deep Neural Networks

引用

IEEE/CAA Journal of Automatica Sinica 2021年第12期8卷 1964-1976页

作者： Tianhao Zhang Jiuhong Xiao Liang Li Chen Wang Guangming Xie State Key Laboratory of Turbulence and Complex Systems Intelligent Biomimetic Design LaboratoryCollege of EngineeringPeking UniversityBeijing 100871China National Engineering Research Center of Software Engineering Peking UniversityBeijing 100871 Institute of Ocean Research Peking UniversityBeijing100871 Peng Cheng Laboratory Shenzhen 518055China

Controlling multiple multi-joint fish-like robots has long captivated the attention of engineers and biologists,for which a fundamental but challenging topic is to robustly track the postures of the individuals in real *** requires detecting multiple robots,estimating multi-joint postures,and tracking identities,as well as processing fast in real *** the best of our knowledge,this challenge has not been tackled in the previous *** this paper,to precisely track the planar postures of multiple swimming multi-joint fish-like robots in real time,we propose a novel deep neural network-based method,named *** TAB part fuses the top-down and bottom-up approaches for vision-based pose estimation,while the IOL part with long short-term memory considers the motion constraints among joints for precise pose *** satisfying performance of our TAB-IOL is verified by testing on a group of freely swimming fish-like robots in various scenarios with strong disturbances and by a deed comparison of accuracy,speed,and robustness with most state-of-the-art ***,based on the precise pose estimation and tracking realized by our TAB-IOL,several formation control experiments are conducted for the group of fish-like *** results clearly demonstrate that our TAB-IOL lays a solid foundation for the coordination control of multiple fish-like robots in a real working *** believe our proposed method will facilitate the growth and development of related fields.

关键词： Deep neural networks formation control multiple fish-like robots pose estimation pose tracking

来源：评论

学校读者我要写书评

暂无评论

Pipeline Anomaly Detection Based on KPCA and Cosine Distance Prototypical Network

Pipeline Anomaly Detection Based on KPCA and Cosine Distance...

引用

IEEE International Conference on Industrial Technology (ICIT)

作者： Kang Yi Xin Liu Wenkai Hu Yupeng Li School of Automation China University of Geosciences Wuhan China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan China

ISBN: (数字)9798331521950

ISBN: (纸本)9798331521967

Accurately and promptly detecting the pipeline anomaly is crucial to the safe operation of pipeline systems, while a difficulty lies in that many existing methods require massive data for training models. However, pipelines are running under normal state for the most of the time, and labeled pipeline anomaly data is usually scarce. Among the commonly used sensors, vibration sensors are widely utilized in pipeline detection because of their advantages such as easy installation and high sensitivity. However, the vibration signal shows non-stationary characteristics when anomalies occur, and are contaminated by noises, making it difficult to represent the actual state with features extracted from either the time or frequency domain. Accordingly, this paper proposes a pipeline anomaly detection method based on the KPCA (kernel principal component analysis) and cosine distance prototypical network. First, features are extracted from original signals; then, the feature dimension is reduced by KPCA; last, the cosine distance is introduced to the prototypical network for anomaly detection. The effectiveness of the proposed method is demonstrated by case studies involving experimental data.

关键词： Vibrations Training Time-frequency analysis Sensitivity Pipelines Sensor phenomena and characterization Feature extraction Time-domain analysis Anomaly detection Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Frequency Domain-Based Low-Rank Approximation for Magnetic Anomaly Detection

Frequency Domain-Based Low-Rank Approximation for Magnetic A...

引用

Sensing, Measurement & Data Analytics in the era of Artificial Intelligence (ICSMD), International Conference on

作者： Yumeng Lv Congyu Liao Huan Liu Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems School of Automation China University of Geosciences Wuhan China Department of Radiology Stanford University CA USA

ISBN: (数字)9798331529192

ISBN: (纸本)9798331529208

Current magnetic anomaly detection (MAD) methods prioritize signal-to-noise ratio (SNR) over signal features and edge information, leading to signal distortion. To address this issue, a new MAD approach utilizes structured lowrank approximation and block singular value decomposition based on the spatial frequency domain, dubbed FL-BSVD is proposed. First, the low-rankness structure of the magnetic anomaly signal is obtained through 2D Discrete Fourier Transform (2D-DFT) and structured Hankel transformation. Then, block singular value decomposition is applied to the Hankel matrix to reduce noise while preserving more signal edge features and enhancing detection accuracy. Finally, a field experiment comparing FL-BSVD with four commonly used methods is conducted. The experiment confirms that FL-BSVD can effectively recover magnetic anomaly signal features and edge information in a strong noisy environment.

关键词： Image edge detection Frequency-domain analysis Noise reduction Magnetic domains Feature extraction Noise measurement Magnetic anomaly detection Signal to noise ratio Singular value decomposition Periodic structures

来源：评论

学校读者我要写书评

暂无评论

From Fourier to neural ODEs: flow matching for modeling complex systems 24

From Fourier to neural ODEs: flow matching for modeling comp...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Xin Li Jingdong Zhang Qunxi Zhu Chengli Zhao Xue Zhang Xiaojun Duan Wei Lin College of Science National University of Defense Technology Changsha Hunan China School of Mathematical Sciences LMNS and SCMS and Research Institute of Intelligent Complex Systems Fudan University China Research Institute of Intelligent Complex Systems Fudan University China School of Mathematical Sciences LMNS and SCMS and Research Institute of Intelligent Complex Systems and State Key Laboratory of Medical Neurobiology and MOE Frontiers Center for Brain Science Institutes of Brain Science Fudan University China and Shanghai Artificial Intelligence Laboratory China

Modeling complex systems using standard neural ordinary differential equations (NODEs) often faces some essential challenges, including high computational costs and susceptibility to local optima. To address these challenges, we propose a simulation-free framework, called Fourier NODEs (FNODEs), that effectively trains NODEs by directly matching the target vector field based on Fourier analysis. Specifically, we employ the Fourier analysis to estimate temporal and potential high-order spatial gradients from noisy observational data. We then incorporate the estimated spatial gradients as additional inputs to a neural network. Furthermore, we utilize the estimated temporal gradient as the optimization objective for the output of the neural network. Later, the trained neural network generates more data points through an ODE solver without participating in the computational graph, facilitating more accurate estimations of gradients based on Fourier analysis. These two steps form a positive feedback loop, enabling accurate dynamics modeling in our framework. Consequently, our approach outperforms state-of-the-art methods in terms of training time, dynamics prediction, and robustness. Finally, we demonstrate the superior performance of our framework using a number of representative complex systems.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Inverse Compensation Feedforward Control Strategy for a Vertical Pneumatic Artificial Muscle System 8

Inverse Compensation Feedforward Control Strategy for a Vert...

引用

8th International Conference on Control Science and systems Engineering, ICCSSE 2022

作者： Xiao, Huai Meng, Qingxin Lai, Xuzhi Yan, Ze Zhao, Shiying Wu, Min School of Automation China University of Geosciences Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Hubei Wuhan430074 China

ISBN: (纸本)9781665484138

This paper proposes an inverse compensation feed-forward control strategy for a vertical pneumatic artificial muscle (PAM) system. Firstly, we conduct a preliminary experiment on the vertical PAM system, and on this basis, we analyze the motion characteristics of the system. Moreover, we clarify the control objective of this paper. Then, we establish a model that can describe the hysteresis characteristics of the system, and we further construct an inverse compensation feedforward controller by inversing this established model. Finally, we carry out some tracking control experiments based on the vertical PAM experimental platform to verify the effectiveness and superiority of the proposed control strategy, © 2022 IEEE.

关键词： Hysteresis

来源：评论

学校读者我要写书评

暂无评论

Recursive estimation with compensation strategies under random access protocol and deception attacks

引用

IEEE Transactions on Signal Processing 2025年 73卷 1954-1965页

作者： Li, Jiaxing Caballero-Aguila, Raquel Hu, Jun Linares-Perez, Josefa Harbin University of Science and Technology Department of Applied Mathematics Harbin150080 China University of Jaén Department of Statistics and Operations Research Jaén23071 Spain Harbin University of Science and Technology Heilongjiang Provincial Key Laboratory of Optimization Control and Intelligent Analysis for Complex Systems Harbin150080 China Harbin University of Science and Technology School of Automation Harbin150080 China University of Granada Department of Statistics and Operations Research Granada18071 Spain

In this paper, recursive least-squares linear estimation algorithms are proposed for stochastic systems influenced by uniform quantization, random access protocol (RAP) and deception attacks. With the purpose of enhancing communication efficiency and reducing unnecessary data collisions, RAP is adopted to schedule data signal transmissions that are also subject to deception attacks. In order to alleviate the side effect of missing information caused by RAP, three compensation strategies (zero-input, zero-order hold and prediction-compensation) are utilized. By resorting to an innovation method, covariance-based filters are designed and then fixed-point smoothers are obtained in light of available observations. Finally, a simulation experiment with comparisons is employed to demonstrate the effectiveness of the developed recursive estimation schemes, where the influence of attack probabilities on estimation accuracy is evaluated. © 1991-2012 IEEE.

关键词： Least squares approximations

来源：评论

学校读者我要写书评

暂无评论

Long‐time target tracking algorithm based on re‐detection multi‐feature fusion

引用

IET Cyber-systems and Robotics 2022年第1期4卷 38-50页

作者： Junsuo Qu Chenxue Tang Yuan Zhang Kai Zhou Abolfazl Razi School of Automation Xi'an Key Laboratory of Advanced Control and Intelligent ProcessXi'an University of Posts and TelecommunicationsXi'anShaanxiChina School of Communication and Information Engineering Xi'an University of Posts and TelecommunicationsXi'anShaanxiChina School of Informatics Computing and Cyber SystemsNorthern Arizona UniversityFlagstaffAZUSA

This paper considers the problem of long-term target tracking in complex scenes when tracking failures are unavoidable due to illumination change,target deformation,scale change,motion blur,and other *** specifically,a target tracking algorithm,called re-detection multi-feature fusion,is proposed based on the fusion of scale-adaptive kernel correlation filtering and *** target tracking algorithm trains three kernel correlation filters based on the histogram of oriented gradients,colour name,and local binary pattern features and then obtains the fusion weight of response graphs corresponding to different features based on average peak correlation energy criterion and uses weighted average to complete the position estimation of the tracked *** order to deal with the problem that the target is occluded and disappears in the tracking process,a random fern classifier is trained to perform re-detection when the target is *** comparing the OTB-50 target tracking dataset,the experimental results show that the proposed tracker can track the target well in the occlusion attribute video sequence in the OTB-100 test dataset and has a certain improvement in tracking accuracy and success rate compared with the traditional correlation filter tracker.

关键词： machine learning pedestrian identification robustness visual surveillance visual tracking

来源：评论

学校读者我要写书评

暂无评论

Optimal Strategies for the Game with Two Faster 3D Pursuers and One Slower 2D Evader 41

Optimal Strategies for the Game with Two Faster 3D Pursuers ...

引用

第41届中国控制会议

作者： Runqi Zhang Shuai Li Chen Wang Guangming Xie State Key Laboratory of Turbulence and Complex Systems Intelligent Biomimetic Design LabCollege of EngineeringPeking University National Engineering Research Center of Software Engineering Peking University Center for Multi-Agent Research Institute for Artificial IntelligencePeking University

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

A pursuit-evasion differential game with two faster pursuers in 3 D space and one slower evader in a 2 D plane is *** pursuers,which have different speeds,attempt to cooperate to capture the evader while the evader struggles to avoid or delay the *** differential games and the concept of isochron,we present a complete solution to this game and develop the optimal strategies for both the pursuers and the *** order to guarantee the optimality of the proposed strategies,the value function of the game is given and is further proved to be the solution of the corresponding Hamilton-Jacobi-Isaacs ***,simulations are carried out with comparisons between the proposed strategies and other typical strategies,the results of which show the optimality of the proposed strategies.

关键词： Pursuit-evasion differential games multi-agent system optimal strategy

来源：评论

学校读者我要写书评

暂无评论

SETTP: Style Extraction and Tunable Inference via Dual-Level Transferable Prompt Learning 27

SETTP: Style Extraction and Tunable Inference via Dual-Level...

引用

27th European Conference on Artificial Intelligence, ECAI 2024

作者： Jin, Chunzhen Huang, Yongfeng Wang, Yaqi Cao, Peng Zaïane, Osmar Northeastern University Shenyang China Key Laboratory of Intelligent Computing in Medical Image of Ministry of Education Northeastern University Shenyang China Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong National Frontiers Science Center for Industrial Intelligence and Systems Optimization Shenyang China Amii University of Alberta EdmontonAB Canada

ISBN: (纸本)9781643685489

Text style transfer, an important research direction in natural language processing, aims to adapt the text to various preferences but often faces challenges with limited resources. In this work, we introduce a novel method termed Style Extraction and Tunable Inference via Dual-level Transferable Prompt Learning (SETTP) for effective style transfer in low-resource scenarios. First, SETTP learns source style-level prompts containing fundamental style characteristics from high-resource style transfer. During training, the source style-level prompts are transferred through an attention module to derive a target style-level prompt for beneficial knowledge provision in low-resource style transfer. Additionally, we propose instance-level prompts obtained by clustering the target resources based on the semantic content to reduce semantic bias. We also propose an automated evaluation approach of style similarity based on alignment with human evaluations using ChatGPT-4. Our experiments across three resourceful styles show that SETTP requires only 1/20th of the data volume to achieve performance comparable to state-of-the-art methods. In tasks involving scarce data like writing style and role style, SETTP outperforms previous methods by 16.24%. © 2024 The Authors.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Blast Furnace Gas Utilization Rate Prediction Model based on Northern Goshawk Optimization and Long Short-Term Memory in Massive Data Set

Blast Furnace Gas Utilization Rate Prediction Model based on...

引用

Industrial Cyber-Physical systems (ICPS)

作者： Yue Zhou Weihua Cao Zhuofu Zhang Yan Yuan School of Automation China University of Geosciences Wuhan China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan China

Affected by the global New Crown Pneumonia epidemic, energy prices in the global market continue to rise. The high quality production and energy saving of steel companies have become an important task. The blast furnace is the front-end core of the steel manufacturing process. The blast furnace gas utilization rate can effectively characterize the internal airflow distribution, blast furnace operation and energy consumption level of the blast furnace. The blast furnace Gas Utilization Rate (GUR) can effectively characterize the internal airflow distribution, blast furnace operation and energy consumption level. Most of the existing studies are based on data-driven models. The shallow neural network model is selected. But blast furnace iron making has complex uncertainty. Massive data samples under industrial information technology need to be processed. The robustness and generalization ability of the prediction model are not satisfactory. To address the above problems, this paper proposes a prediction model based on NGO-LSTM regression. The model parameter searchs can be intelligently. It achieves high-precision prediction results for massive data samples. Firstly, the selection of feature parameters is completed by maximal information coefficient. Secondly, the strong coupling of blast furnace ground needs to fully predict the relationship between the characteristic parameters. The Long Short-Term Memory (LSTM) neural network with certain memory capability is selected. And multiple parameters of this neural network model are optimized by the Northern Goshawk Optimization (NGO) algorithm. An NGO-LSTM regression prediction model is established. In this paper, experiments are carried out using actual production data. The experimental results show that the proposed method can accurately predict the blast furnace gas utilization rate. This can provide a reference for improving blast furnace product quality, reducing costs and increasing efficiency.

关键词： Energy consumption Neural networks Production Predictive models Blast furnaces Prediction algorithms Robustness

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：