检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Wang, Guangming Hu, Yunzhe Liu, Zhe Zhou, Yiyang Tomizuka, Masayoshi Zhan, Wei Wang, Hesheng Department of Automation Key Laboratory of System Control and Information Processing of Ministry of Education Key Laboratory of Marine Intelligent Equipment and System of Ministry of Education Shanghai Jiao Tong University China MoE Key Lab of Artificial Intelligence Shanghai Jiao Tong University China Mechanical Systems Control Laboratory University of California Berkeley United States

3D scene flow estimation from point clouds is a low-level 3D motion perception task in computer vision. Flow embedding is a commonly used technique in scene flow estimation, and it encodes the point motion between two consecutive frames. Thus, it is critical for the flow embeddings to capture the correct overall direction of the motion. However, previous works only search locally to determine a soft correspondence, ignoring the distant points that turn out to be the actual matching ones. In addition, the estimated correspondence is usually from the forward direction of the adjacent point clouds, and may not be consistent with the estimated correspondence acquired from the backward direction. To tackle these problems, we propose a novel all-to-all flow embedding layer with backward reliability validation during the initial scene flow estimation. Besides, we investigate and compare several design choices in key components of the 3D scene flow network, including the point similarity calculation, input elements of predictor, and predictor & refinement level design. After carefully choosing the most effective designs, we are able to present a model that achieves the state-of-the-art performance on FlyingThings3D and KITTI Scene Flow datasets. Our proposed model surpasses all existing methods by at least 38.2% on FlyingThings3D dataset and 24.7% on KITTI Scene Flow dataset for EPE3D metric. We release our codes at https://***/IRMVlab/3DFlow. Copyright © 2022, The Authors. All rights reserved.

关键词： Embeddings

来源：评论

学校读者我要写书评

暂无评论

FCS-MPC for a single-phase two-stage grid-connected PV inverter

引用

IET POWER ELECTRONICS 2019年第4期12卷 915-922页

作者： Liu, Jiang Cheng, Shanmei Liu, Ying Shen, Anwen Huazhong Univ Sci & Technol Sch Automat Key Lab Image Proc & Intelligent Control Wuhan Hubei Peoples R China

The high-efficiency inverter in a wide input voltage range is expected to be a solution for the distributed generation system. This study gives a high-efficiency two-stage topology and proposes a finite-control-set model predictive control (FCS-MPC) strategy. The main objectives of the proposed model predictive controller are: (i) simultaneously controlling multi-variables in the complex buck-boost current controller, (ii) optimal design calculation for predictive current and voltage, (iii) normalisation of the state variables for cost function formulation, and (iv) low correlation between the dc-link capacitor and control algorithm. A more reasonable operation mode of high-frequency switches is provided for the two-stage topology improving system performance. In order to optimise the total harmonic distortion of the output current and the error of the dc-link voltage, a suitable weighting factor could be designed in the proposed FCS-MPC method. The steady-state characteristics of the proposed controller are simulated based on MATlab/Simulink. Experimental results demonstrate the validity and practicability of the proposed control strategy.

关键词： power convertors electric current control voltage control photovoltaic power systems invertors power grids predictive control harmonic distortion main objectives model predictive controller multivariables complex buck-boost current controller optimal design calculation predictive current voltage state variables cost function formulation dc-link capacitor control algorithm high-frequency switches system performance output current dc-link voltage FCS-MPC method control strategy single-phase two-stage grid high-efficiency inverter wide input voltage range distributed generation system high-efficiency two-stage topology finite-control-set model predictive control

来源：评论

学校读者我要写书评

暂无评论

Online Attentive Kernel-Based Temporal Difference Learning

arXiv

引用

arXiv 2022年

作者： Yang, Guang Chen, Xingguo Yang, Shangdong Wang, Huihui Dong, Shaokang Gao, Yang The the Jiangsu Key Laboratory of Big Data Security & Intelligent Processing Nanjing University of Posts and Telecommunications National Engineering Laboratory for Agri-Product Quality Traceability Beijing Technology and Business University China The State Key Laboratory for Novel Software Technology Nanjing University China The PCA Lab Key Lab of Intelligent Perception and Systems for High-Dimensional Information Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security School of Computer Science and Engineering Nanjing University of Science and Technology China

With rising uncertainty in the real world, online Reinforcement Learning (RL) has been receiving increasing attention due to its fast learning capability and improving data efficiency. However, online RL often suffers from complex Value Function Approximation (VFA) and catastrophic interference, creating difficulty for the deep neural network to be applied to an online RL algorithm in a fully online setting. Therefore, a simpler and more adaptive approach is introduced to evaluate value function with the kernel-based model. Sparse representations are superior at handling interference, indicating that competitive sparse representations should be learnable, non-prior, non-truncated and explicit when compared with current sparse representation methods. Moreover, in learning sparse representations, attention mechanisms are utilized to represent the degree of sparsification, and a smooth attentive function is introduced into the kernel-based VFA. In this paper, we propose an Online Attentive Kernel-Based Temporal Difference (OAKTD) algorithm using two-timescale optimization and provide convergence analysis of our proposed algorithm. Experimental evaluations showed that OAKTD outperformed several Online Kernel-based Temporal Difference (OKTD) learning algorithms in addition to the Temporal Difference (TD) learning algorithm with Tile Coding on public Mountain Car, Acrobot, CartPole and Puddle World tasks. Copyright © 2022, The Authors. All rights reserved.

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Deployment of second-order networked mobile agents over a smooth curve

引用

AUTOMATICA 2022年 146卷

作者： Su, Housheng Xu, Qiuju Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Image Proc & Intelligent Control Key Lab Educ Minist China Luoyu Rd 1037 Wuhan 430074 Peoples R China Chongqing Normal Univ Sch Math Sci Univ Town Middle Rd 37 Chongqing 401331 Peoples R China

In this paper, we investigate second-order deployment of multiple mobile agents over a C2 curve in 2D or 3D space. We assume that an informed agent in the group can measure its relative position with respect to the given curve and velocity, other agents have access to the local information about the given curve as well as the position and velocity in regard to their closest neighbors, where the agent dynamics is described by a strongly damped wave equation. Without delay, we establish a boundary control law for the informed agent. When taking into account time-varying delay, we assume that the informed agent transmits its measurements to its neighbors through a communication mechanism. Given appropriate controller gains, second-order deployment can be achieved with an exponential decay. Finally, some simulation results are presented to illustrate our theoretical results.(c) 2022 Elsevier Ltd. All rights reserved.

关键词： Wave equations

来源：评论

学校读者我要写书评

暂无评论

Bit cyclic shift method to reinforce digital chaotic maps and its application in pseudorandom number generator

引用

APPLIED MATHEMATICS AND COMPUTATION 2022年 420卷 126788-126788页

作者： Zheng, Jun Hu, Hanping Huazhong Univ Sci & Technol Sch Cyber Sci & Engn Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China Minist Educ Key Lab Image Informat Proc & Intelligent Control Wuhan 430074 Peoples R China

When chaotic systems are realized on finite precision devices, dynamical degradation occurs, which causes many flaws in chaos-based applications. In this paper, a new bit cyclic shift method is introduced to reinforce digital chaotic maps and counteract the problem of dynamical degradation. The state value of digital chaotic map is modified by circularly shifting its bits according to a designed shift strategy. This method is applied to the Logistic map to generate many new improved chaotic maps by changing the shift strategy. Theoretical analysis and numerical experiments demonstrate their chaotic parameter ranges and chaotic behaviors. Furthermore, the simulation results under low computing precision indicate that our method has the ability to counteract the dynamical degradation. This method has high flexibility and can apply to any chaotic system. Finally, a new pseudorandom number generator (PRNG) is designed to show the good performance of the improved digital chaotic maps.(c) 2021 Elsevier Inc. All rights reserved.

关键词： Digital chaotic maps Finite precision Dynamical degradation Bit shift Pseudorandom number generator

来源：评论

学校读者我要写书评

暂无评论

Research on Public Opinion Guidance Strategy based on Interaction Constraints 5

Research on Public Opinion Guidance Strategy based on Intera...

引用

IEEE 5th Information Technology and Mechatronics Engineering Conference (ITOEC)

作者： Xu, Yuanyuan Wu, Zhan Chen, Xi Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan Peoples R China Educ Minist China Image Proc & Intelligent Control Key Lab Wuhan Peoples R China

ISBN: (纸本)9781728143231;9781728143224

The dissemination of public opinion has a considerable impact on people's social lives, and the question of how both the government and the media can correctly guide public opinion and restore social stability is an issue that requires urgent attention. Combining the characteristics of public opinion dissemination in the network society leads to individual interaction. Considering the interaction between individual views and network topology, this paper combines the collaborative network model with the distance-constrained Hegselmann-Krause (HK) model to propose a consensus-guided strategy based on interaction constraints. It does this by changing the network topology and limiting the scope of individual interactions, thereby affecting the spread of public opinion. Based on the small-world network, the collaborative network model introduces the two parameters of Connection edge probability p and broken edge probability q, while also proposing a collaborative network model for Connection edge's breaking of the network based on the difference of individual opinion values. This paper designs simulation experiments by changing p, q, and the interaction distance, while also analyzing the results of public opinion's evolution from the perspectives of timeliness and the diversity of viewpoints. The experimental results show that when individuals are restricted from interacting with the surrounding environment and the scope of interaction between individuals is restricted, this will slow down the evolution of public opinion and increase the number of opinion clusters. By contrast, when it is encouraged, the interaction between individuals and the surrounding environment will make the entire system communicate more effectively within a short period of time, promoting the evolutionary benefits of public opinion considerably, shortening the convergence time of public opinion evolution to reach a steady state, and reducing the number of opinion clusters. The strategy is consist

关键词： evolutionary dynamics of public opinion physical distance interaction constraints social networks

来源：评论

学校读者我要写书评

暂无评论

Despecking of Ultrasound image Using LENet-based Nonlocal-Means Method

Despecking of Ultrasound Image Using LENet-based Nonlocal-Me...

引用

Conference on Medical Imaging - Ultrasonic Imaging and Tomography

作者： Yu Houqiang Li Ling Zheng Min Qiu Wu Ding Mingyue Hubei Univ Sci & Technol Dept Math & Stat Xianning 437100 Hubei Peoples R China Hubei Univ Sci & Technol Xianning Med Coll Sch Biomed Engn & Imaging Xianning 437100 Hubei Peoples R China Hubei Univ Sci & Technol Sch Elect & Informat Engn Xianning 437100 Hubei Peoples R China Huazhong Univ Sci & Technol Educ Minist China Image Proc & Intelligence Control Key Lab Dept Biomed EngnColl Life Sci & Technol Wuhan 430074 Hubei Peoples R China Wuhan Seekmore Intelligent Imaging Inc Wuhan Peoples R China Hubei Yueming Technol Co Ltd Wuhan Peoples R China

ISBN: (纸本)9781510671690;9781510671683

Speckle noise is an integral component in medical ultrasound imaging, which have a random granular pattern formation. This noise degrades the visual quality of ultrasound images and complicates image-based interpretation and diagnosis. The removal of interference-induced noise is a primary challenge as ultrasound image studies seek to achieve higher accuracy and characterize more subtle small and low-contrast lesions. In this study, a novel method, which combines the nonlocal-means (NLM) filter with a simple unsupervised deep model named Laplacian Eigenmaps network (LENet), has been proposed for ultrasonic speckle reduction. The proposed method exploits both the global features, redundancy information and self-similarity properties of noisy images, which first extract features from the noisy image by the Laplacian Eigenmaps algorithm, and then apply it to refine the image self-similarities weight for helping the NLM filter to provide better despeckling performance. Specifically, this is a two-stage approach that the first stage is to learn filter banks from a small quantity of training samples by LENet network and the following stage is to utilize the output eigenvectors as similarity metrics of pixels within the NLM filter. The performance of our approach is compared with related state-of-the-art methods on synthetic images, simulated image and real ultrasound images. The results show that our method can provide better noise removal ability over many previously despecking filters.

关键词： ultrasound image nonlocal-means Laplacian Eigenmaps network despeckling

来源：评论

学校读者我要写书评

暂无评论

Group controllability of two-time-scale discrete-time multi-agent systems

引用

JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS 2020年第6期357卷 3524-3540页

作者： Long, Mingkang Su, Housheng Liu, Bo Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Image Proc & Intelligent Control Key Lab Educ Minist China Luoyu Rd 1037 Wuhan 430074 Peoples R China North China Univ Technol Coll Sci Beijing 100144 Peoples R China

In this paper, we address the group controllability issue for discrete-time multi-agent systems (MASs) with two-time-scale feature. Since the existence of singular perturbation parameter in modeling the two-time-scale MASs will cause ill-posedness problems, we decouple the MASs into fast and slow subsystems by using the iterative and approximate methods, and then eliminate the singular perturbation parameter. By using matrix theory, we propose a lot of necessary and/or sufficient conditions of group controllability, where some easy-to-use criteria are determined only by eigenvalues of matrices. Lastly, a simulation example is given to illustrate the effectiveness of the proposed theoretical results. (C) 2019 The Franklin Institute. Published by Elsevier Ltd. All rights reserved.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Decoupled and memory-reinforced networks: Towards effective feature learning for one-step person search

arXiv

引用

arXiv 2021年

作者： Han, Chuchu Zheng, Zhedong Gao, Changxin Sang, Nong Yang, Yi Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China ReLER University of Technology Sydney Australia Baidu Research China

The goal of person search is to localize and match query persons from scene images. For high efficiency, one-step methods have been developed to jointly handle the pedestrian detection and identification sub-tasks using a single network. There are two major challenges in the current one-step approaches. One is the mutual interference between the optimization objectives of multiple sub-tasks. The other is the sub-optimal identification feature learning caused by small batch size when end-to-end training. To overcome these problems, we propose a decoupled and memory-reinforced network (DMRNet). Specifically, to reconcile the conflicts of multiple objectives, we simplify the standard tightly coupled pipelines and establish a deeply decoupled multi-task learning framework. Further, we build a memory-reinforced mechanism to boost the identification feature learning. By queuing the identification features of recently accessed instances into a memory bank, the mechanism augments the similarity pair construction for pairwise metric learning. For better encoding consistency of the stored features, a slow-moving average of the network is applied for extracting these features. In this way, the dual networks reinforce each other and converge to robust solution states. Experimentally, the proposed method obtains 93.2% and 46.9% mAP on CUHK-SYSU and PRW datasets, which exceeds all the existing one-step methods. Copyright © 2021, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Tight Formulations of Unit Commitment with General Start-Up Ramp Rate Constraints

Tight Formulations of Unit Commitment with General Start-Up ...

引用

Sustainable Power and Energy Conference (iSPEC)

作者： Yijie Gao Xuan Zhang Yaowen Yu Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China Hitachi ABB Power Grid San Jose California US

ISBN: (纸本)9781665414401

Unit commitment (UC) is a central market clearing process in wholesale electricity markets. With the increasing size and complexity of market-clearing models, UC problems become more and more complicated. To improve the computation efficiency, this paper tightens UC formulations with the general start-up ramp rate. Through analyzing their parameters over three periods, units are categorized into several categories for constructing corresponding tight formulations, and three of the categories are beyond existing works. By applying a recent systematic formulation tightening approach on top of these categories, new tight constraints of the general start-up ramp rate have been obtained. Numeral testing results illustrate the tightening process and demonstrate the improvement of computational efficiency when using our tightened formulations.

关键词： Systematics Computational modeling Conferences Electricity supply industry Computational efficiency Complexity theory Testing

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：