检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Wang, Zhong Ren, Lele Wen, Yue Wang, Hesheng The Department of Automation School of Electronics Information and Electrical Engineering Shanghai Jiao Tong University Shanghai200240 China The School of Information and Control Engineering China University of Mining and Technology Xuzhou221008 China

Recent advancements in LiDAR-Inertial Odometry (LIO) have boosted a large amount of applications. However, traditional LIO systems tend to focus more on localization rather than mapping, with maps consisting mostly of sparse geometric elements, which is not ideal for downstream tasks. Recent emerging neural field technology has great potential in dense mapping, but pure LiDAR mapping is difficult to work on high-dynamic vehicles. To mitigate this challenge, we present a new solution that tightly couples geometric kinematics with neural fields to enhance simultaneous state estimation and dense mapping capabilities. We propose both semi-coupled and tightly coupled Kinematic-Neural LIO (KN-LIO) systems that leverage online SDF decoding and iterated error-state Kalman filtering to fuse laser and inertial data. Our KN-LIO minimizes information loss and improves accuracy in state estimation, while also accommodating asynchronous multi-LiDAR inputs. Evaluations on diverse high-dynamic datasets demonstrate that our KN-LIO achieves performance on par with or superior to existing state-of-the-art solutions in pose estimation and offers improved dense mapping accuracy over pure LiDAR-based methods. The relevant code and datasets will be made available at https://**. Copyright © 2025, The Authors. All rights reserved.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Robust control Performance in Uncertain Environments: A Semi-Heuristic Gradient Descent Approach

Robust Control Performance in Uncertain Environments: A Semi...

引用

International Conference on Instrumentation control and automation (ICA)

作者： Mohammed Elbadri Ahmed Hassan Ahmad Irham Jambak Ismail Bayezit Mehmet Turan Soylemez Department of Control and Automation Engineering Istanbul Technical University Istanbul Turkiye Department of Mechatronics Engineering Istanbul Technical University Istanbul Turkiye Department of Aeronautical Engineering Istanbul Technical University Istanbul Turkiye

Designing a robust controller for systems with parameter uncertainties is a complex and demanding task. Traditional deterministic and probabilistic approaches may fall short in providing efficient and satisfactory solutions. To address this challenge, we propose a semi-heuristic approach that exploits the Kharitonov theorem to establish an initial point for the gradient descent algorithm. Through an iterative optimization process that incorporates user-defined performance criteria, our approach provide a robust controller with respect to presence of system uncertainties. Numerical simulations validate the effectiveness of our proposed method and highlight its superiority in addressing robust control problems compared with auto-tuned ΡΠ) controller.

关键词：

来源：评论

学校读者我要写书评

暂无评论

On Uniform Observability of Linear Time-varying Systems with respect to Subspaces and Its Applications

引用

IEEE Transactions on Automatic control 2025年

作者： Liu, Tao Huang, Jie Southern University of Science and Technology Shenzhen Key Laboratory of Control Theory and Intelligent Systems Shenzhen518055 China The Chinese University of Hong Kong Department of Mechanical and Automation Engineering Shatin Hong Kong

Recently, the notion of uniform observability with respect to a subspace was proposed in [21] and was utilized to solve the synchronization problem for single-integrator multi-agent systems with time-varying matrix-weighted couplings. This paper further explores the concept of uniform observability of a linear time-varying system with respect to a subspace. We will first establish the invariance property of uniform observability with respect to some invariant subspace under time-varying output feedback control. Then, based on this invariance property, we will establish an exponential convergence result with respect to an invariant subspace for a class of linear switched systems. Leveraging this exponential convergence result, we will solve the exponential synchronization problem for general linear multi-agent systems over jointly connected switching networks. It is noted that the exponential synchronization problem was also studied recently by some more sophisticated methods. Compared with these methods, our method is much simpler. Moreover, our method leads to an explicit characterization of the exponential synchronization rate. © 1963-2012 IEEE.

关键词： Linear time-invariant system

来源：评论

学校读者我要写书评

暂无评论

Dynamics of coupled implicit and explicit opinions

arXiv

引用

arXiv 2023年

作者： Zhang, Qi 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 Department of Automation School of Mechatronic Engineering and Automation Shanghai University Shanghai200444 China

This paper proposes a dual opinions co-evolution model based on the dual attitudes theory in social psychology, where every individual has dual opinions of an object: implicit and explicit opinions. The implicit opinions are individuals’ inner evaluations affected by the experience, while the explicit opinions are external expressions of the evaluations and are influenced by public opinion. We theoretically analyze the dynamics of coupled dual opinions in an adversarial discussion scenario, in which two opposing groups participate in the discussion. We rigorously give a method to grade the degree of individual bias. When individuals have a slight bias, dual opinions reach a consensus, showing the phenomenon of acceptance. When individuals are moderately or even extremely biased, they will exhibit the phenomenon of outward compliance, that is, their publicly explicit opinions conform to public opinion, while implicit opinions will reach a persistent disagreement and even polarization. Our investigation indicates that dual opinions may not be monotone initially but eventually converge monotonically. Besides, we analyze the influence of parameters on the co-evolutionary process and results. Numerical simulations of the dual opinions formation process produce inward and outward conformity phenomena, namely acceptance and compliance. Copyright © 2023, The Authors. All rights reserved.

关键词： Dynamics

来源：评论

学校读者我要写书评

暂无评论

Small defect detection on the surface of automobile differential housing based on FSSD and large receptive field

Small defect detection on the surface of automobile differen...

引用

2022 Chinese automation Congress, CAC 2022

作者： Chen, Tehuan Wang, Guodong Wang, Pengkai Cui, Jie Shanghai Jiao Tong University Department of Automation Shanghai China Ningbo University Department of Mechanical Engineering Ningbo China Zhejiang University College of Control Science and Engineering Hangzhou China Zhejiang Huashuo Technology Co. Ltd. Ningbo China

ISBN: (纸本)9781665465335

Due to the limited number of identifiable features that can be extracted, existing methods for detecting small defects on the surface of automobile differential housings have difficulty balancing detection speed and accuracy. This paper proposes a lightweight feature fusion single multibox detector with a larger receptive field (RFSSD) algorithm to improve the performance of small defect detection. Firstly, the model replaces VGG16, the backbone network of the feature fusion single shot multibox detector (FSSD) algorithm, with a lightweight MobileNet network to create a more practical feature fusion structure consisting of depthwise separable convolution. Subsequently, the improved receptive field block (RFB) module is introduced into the RFSSD model's shallow layer of a feature pyramid structure. Finally, the size of the default box is modified for the target defect's actual size. According to the experimental results, the mAP of the RFSSD model is 96.7 percent, up 12.4 percent from before the improvement. The detection speed of 158FPS and the model memory of 25.3MB outperform many state-of-the-art object detection models. Our model can detect various small defects of differential housings surface in real-time and accurately. © 2022 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Day-ahead Unit Commitment Incorperating Deep Peak Regulation 3

Day-ahead Unit Commitment Incorperating Deep Peak Regulation

引用

3rd International Conference on Energy engineering and Power Systems, EEPS 2023

作者： Xiao, Yang Ding, Tao Liu, Yuzheng Zhang, Biyuan Su, Bo Guo, Xiaorui Xi'an Jiaotong University School of Electrical Engineering Xi'an China State Grid Ningxia Electric Power Co. Ltd Center of Electric Power Control Yinchuan China China Electric Power Research Institute Electric Power Automation Department Beijing China

ISBN: (纸本)9798350313857

To give the potential regulation capacity of the thermal power generator for renewable energy integration by peak regulation, this paper has investigated in developing a comprehensive model for day-ahead unit commitment, taking into account deep peak regulation. Specifically, the investigation entails dividing the peak regulation into distinct modes and establishing corresponding operational cost functions. Subsequently, a mixed-integer linear programming model is presented for power scheduling, leveraging the defined cost function. To validate the effectiveness of the proposed model and highlight the superiority of deep peak regulation in accommodating renewable energy sources, a modified IEEE-118 case has been conducted. © 2023 IEEE.

关键词： Cost functions

来源：评论

学校读者我要写书评

暂无评论

Automatic Nonlinear MPC Approximation with Closed-loop Guarantees

引用

IEEE Transactions on Automatic control 2025年

作者： Tokmak, Abdullah Fiedler, Christian Zeilinger, Melanie N. Trimpe, Sebastian Kohler, Johannes Aalto University Department of Electrical Engineering and Automation Espoo02150 Finland RWTH Aachen University Institute for Data Science in Mechanical Engineering Aachen52068 Germany ETH Zurich Institute for Dynamic Systems and Control ZurichCH-8092 Switzerland

Safety guarantees are vital in many control applications, such as robotics. Model predictive control (MPC) provides a constructive framework for controlling safety-critical systems but is limited by its computational complexity. We address this problem by presenting a novel algorithm that automatically computes an explicit approximation to nonlinear MPC schemes while retaining closed-loop guarantees. Specifically, the problem can be reduced to a function approximation problem, which we then tackle by proposing Alkia-x, the Adaptive and Localized Kernel Interpolation Algorithm with eXtrapolated reproducing kernel Hilbert space norm. Alkia-x is a non-iterative algorithm that ensures well-conditioned computations, a fast-to-evaluate approximating function, and the guaranteed satisfaction of any desired bound on the approximation error. Hence, Alkia-x automatically computes an explicit function that approximates the MPC, yielding a controller suitable for safety-critical systems and high sampling rates. We apply Alkia-x to approximate two nonlinear MPC schemes, demonstrating reduced computational demand and applicability to realistic problems. © 1963-2012 IEEE.

关键词： Predictive control systems

来源：评论

学校读者我要写书评

暂无评论

Successive statistical iterative reconstruction, 3D-filtering and region growing methods for high-quality 3D visualization of cone-beam CT image 2022

Successive statistical iterative reconstruction, 3D-filterin...

引用

7th International Conference on Multimedia and Image Processing, ICMIP 2022

作者： Dong, Jian Zhang, Siyuan Yang, Xiaoxia Zhang, Jingyu Tianjin Key Laboratory of Information Sensing and Intelligent Control Tianjin University of Technology Tianjin300222 China Department of Automation and Electrical Engineering Tianjin University of Technology and Education Tianjin300222 China

ISBN: (纸本)9781450387408

Our study is to improve the accuracy of bone morphology depicted in dental cone-beam computed tomography (CBCT) images. First, successive statistical iterative reconstruction was used to reduce unavoidable streak artefacts. Then 3-dimentional filtering (Gaussian-Laplacian filter) and region growing methods were examined to present clear bone morphology. CBCT examinations were performed with the following principal exposure parameters: I-mode, FOV 10cm in diameter, 120 kV, 15 mA, 0.2 mm slice thickness, and exposure time of 10s. Ordered subset-expectation maximization (OS-EM) algorithm was applied for unavoidable streak artefact reduction. 3D Laplacian sharpening to images preprocessed by Gaussian smoothing was sequentially tested. Region growing method with dilation and erosion was used to segment maxillofacial tissue. Streak artefact induced by metallic prosthetic appliances was reduced by applying successive iterative OS-EM algorithm. Multi-planar reconstruction (MPR) images at left side molar plane and mid-sagittal plane were presented to validate effect of OS-EM algorithm and 3D Gaussian-Laplacian filter. Maxillofacial tissue was segmented and presented to show the effect of region growing method. Streak artefact reducing method, 3D filtering method for image smoothing and sharpening, and region growing method with dilation and erosion were effective to improve accuracy of bone morphology in dental CBCT images. © 2022 ACM.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Partially Occluded Face Expression Recognition with CBAM-Based Residual Network for Teaching Scene

Partially Occluded Face Expression Recognition with CBAM-Bas...

引用

2023 China automation Congress, CAC 2023

作者： Bai, Yanxing Chen, Luefeng Li, Min Wu, Min Pedrycz, Witold Hirota, Kaoru School of Automation China University of Geosciences Wuhan430074 China School of Automation Ministry of Education China University of Geosciences Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Wuhan430074 China University of Alberta Department of Electrical and Computer Engineering EdmontonABT6G 2R3 Canada Systems Research Institute Polish Academy of Sciences Warsaw00-901 Poland Istinye University Department of Computer Engineering Sariyer Istanbul34396 Turkey Tokyo Institute of Technology Yokohama226-8502 Japan

ISBN: (纸本)9798350303759

In this paper, a deep residual network based on convolutional block attention module (CBAM) is proposed, which is utilized for feature extraction of partially occluded face expression data. The proposed method overcomes the problem of localized occlusion face feature extraction by focusing on the regions and channels containing important information in the occluded face data through CBAM. Multi-task cascaded convolutional networks (MTCNN) are firstly utilized to localize the key regions of face emotion, and then deep emotion features are extracted by CBAM-ResNet network. The final emotion labels are generated. The effectiveness of this paper's method is verified on the RAF-DB dataset and the occluded CK+ dataset. The experimental accuracy in the RAF-DB dataset is 76.3%, which is 3.74% and 1.64% higher than the accuracy produced by the method of RGBT, and the WLS-RF, respectively. Application experiments are carried out in the real teaching scenario, which verifies the applicability of the algorithm in the real teaching scene. © 2023 IEEE.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

A Knowledge-driven Cross-modal Memory Network for Radiology Report Generation

A Knowledge-driven Cross-modal Memory Network for Radiology ...

引用

2023 CAA Symposium on Fault Detection, Supervision and Safety for Technical Processes, SAFEPROCESS 2023

作者： Fan, Junyan Wang, Jing Zhou, Meng Zhang, Yanzhu Shi, Mingyu North China University of Technology School of Electrical and Control Engineering Beijing China Shenyang Ligong University School of Automation and Electrical Engineering Shenyang China The Fourth Affiliated Hospital of China Medical University Eye Hospital of China Medical University The Key Laboratory of Lens in Liaoning Province Shenyang Department of Ophthalmology China

ISBN: (纸本)9798350337754

In previous studies, most work focused on generating high-quality reports by smoothing the alignment between cross-modal mappings. However, a common problem with current radiology report generation methods is that they ignore the interactions and connections between various medical findings that are critical to medical discovery. To overcome this problem, we propose a knowledge-driven cross-modal memory network. The network is built by data mining clinical reports and constructing knowledge graphs and their corresponding adjacency matrices based on the mined a prior knowledge. Combining information after graphical interaction through a cross-modal memory network with a knowledge graph adjacency matrix. Finally the Transformer generates a full report. Experiments have demonstrated that reports generated from cross-modal memory network models that incorporate prior knowledge are more clinically informative. © 2023 IEEE.

关键词： Knowledge graph

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：