检索结果-内蒙古大学图书馆

Partially Double-Pass Structure in Serial Hybrid Fiber Amplifier

学校读者我要写书评

暂无评论

Partially Double-Pass Structure in Serial Hybrid Fiber Ampli...

engineering, Science and Advanced Technology (ICESAT), International Conference on

作者： Ali Yaseen Ali Mohammed Kamil Salh Al-Mashhadani Thamer Fahad Al-Mashhadani Rabi Noori Hammudi Yassine Bouteraa Control and Energy Management Laboratory (CEM Lab)Ecole Nationaled Ingenieurs de Sfax (ENIS) Institut Superieur de Biotechnologie de Sfax (ISBS) University of Sfax Tunisia Department of Petroleum Systems Control Engineering Tikrit University Saladdin Iraq Department of Electrical Engineering Tikrit University Saladdin Iraq Department of Laser Engineering and optoelectronics Kut Univeristy College Alkut Wasit Iraq Department of Computer Engineering Prince Sattam Bin Abdulaziz University Al-Kharj Saudi Arabia

In this paper, a partially double pass configuration in serial hybrid fiber amplifier is experimentally demonstrated. In the proposed design, a double pass erbium gain and single pass Raman gain are achieved serially. A total pump power of 450 mW (400 mW for 1495 nm Raman amplifier and 50 mW for1480 nm in erbium amplifier) were used. At -30 dBm input signal power and optimum pumps conditions, the achieved flatness bandwidth is 80 nm (1530–1610 nm) in the conventional and long bands (C+L) bands. In addition, the obtained average gain level is 33 dB. While the obtained flatness gain is 85 nm (1525–1610 nm) within the large input signal power region at -5 dBm. By choosing a proper pump wavelength that avoid the overlapping between Raman and erbium peaks gain, a wide flatness gain is obtained.

关键词：

Convergence Analysis of the Best Response Algorithm for Time-Varying Games

学校读者我要写书评

暂无评论

Convergence Analysis of the Best Response Algorithm for Time...

IEEE Conference on Decision and control

作者： Zifan Wang Yi Shen Michael M. Zavlanos Karl H. Johansson Division of Decision and Control Systems School of Electrical Enginnering and Computer Science KTH Royal Institute of Technology and also with Digital Futures Stockholm Sweden Department of Mechanical Engineering and Materials Science Duke University Durham NC USA

This paper studies a class of strongly monotone games involving non-cooperative agents that optimize their own time-varying cost functions. We assume that the agents can observe other agents' historical actions and choose actions that best respond to other agents' previous actions; we call this a best response scheme. We start by analyzing the convergence rate of this best response scheme for standard time-invariant games. Specifically, we provide a sufficient condition on the strong monotonicity parameter of the time-invariant games under which the proposed best response algorithm achieves exponential convergence to the static Nash equilibrium. We further illustrate that this best response algorithm may oscillate when the proposed sufficient condition fails to hold, which indicates that this condition is tight. Next, we analyze this best response algorithm for time-varying games where the cost functions of each agent change over time. Under similar conditions as for time-invariant games, we show that the proposed best response algorithm stays asymptotically close to the evolving equilibrium. We do so by analyzing both the equilibrium tracking error and the dynamic regret. Numerical experiments on economic market problems are presented to validate our analysis.

关键词：

Distract Your Attention: Multi-head Cross Attention Network for Facial Expression Recognition

学校读者我要写书评

暂无评论

arXiv

arXiv 2021年

作者： Wen, Zhengyao Lin, Wenzhong Wang, Tao Xu, Ge Fujian Provincial Key Laboratory of Information Processing and Intelligent Control College of Computer and Control Engineering Minjiang University Fuzhou China

This paper presents a novel facial expression recognition network, called Distract your Attention Network (DAN). Our method is based on two key observations in biological visual perception. Firstly, multiple facial expression classes share inherently similar underlying facial appearance, and their differences could be subtle. Secondly, facial expressions simultaneously exhibit themselves through multiple facial regions, and for recognition, a holistic approach by encoding high-order interactions among local features is required. To address these issues, this work proposes DAN with three key components: Feature Clustering Network (FCN), Multi-head Attention Network (MAN), and Attention Fusion Network (AFN). Specifically, FCN extracts robust features by adopting a large-margin learning objective to maximize class separability. In addition, MAN instantiates a number of attention heads to simultaneously attend to multiple facial areas and build attention maps on these regions. Further, AFN distracts these attentions to multiple locations before fusing the feature maps to a comprehensive one. Extensive experiments on three public datasets (including AffectNet, RAF-DB, and SFEW 2.0) verified that the proposed method consistently achieves state-of-the-art facial expression recognition performance. The DAN code is publicly available. Copyright © 2021, The Authors. All rights reserved.

关键词： Face recognition

Protein-Folding Inspired Programmable control System for Robot Swarms

学校读者我要写书评

暂无评论

Protein-Folding Inspired Programmable Control System for Rob...

IEEE International Conference on Robotics and Biomimetics

作者： Feng Zhang Yuping Gu Zebin Wang Shuai Yuan Yongliang Yang School of Electrical and Control Engineering Shenyang Jianzhu University Shenyang Liaoning China State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Shenyang China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang Liaoning China School of Computer Science and Engineering Shenyang Jianzhu University Shenyang Liaoning China

ISBN: (数字)9798331509644

ISBN: (纸本)9798331509651

In swarm robotics, most of the control algorithms are inspired by the behaviors of animal swarms. The capability gap between robots and animals, however, hinders the reliability of these algorithms. Toward developing reliable algorithms of robot swarms, a promising solution is learning from the collective behaviors of low-intelligent natural swarms, such as cell collectives and molecular assemblies. Here, we proposed a programmable control system inspired by protein folding. Using our homemade RoboFold platform, we demonstrated a protein-folding inspired programmable control method for robot swarms. By programming and designing the sequence of a thread of protein-like robots, the developed system can mimic the formation of hierarchical structures in robot swarms as the three-level of protein structures. We quantitatively characterized the specific features of self-organized morphologies. The experimental results together demonstrated the feasibility of this method. To summarize, this work successfully translated the principles of protein-folding into programming methods to control the morphology of robot swarms. Considering that the protein folding problem is essentially a physicochemical problem, this system is not only beneficial for the control of swarm robots but also has the potential to develop into a tool for studying the protein folding problem.

关键词： Proteins Vibrations Programmable control Animals Morphology Swarm robotics Programming control systems Robustness Robots

Rethinking Guidance Information to Utilize Unlabeled Samples: A Label Encoding Perspective

学校读者我要写书评

暂无评论

arXiv

arXiv 2024年

作者： Zhang, Yulong Yao, Yuan Chen, Shuhao Jin, Pengrong Zhang, Yu Jin, Jian Lu, Jiangang Department of Computer Science and Engineering Southern University of Science and Technology Shenzhen China College of Control Science and Engineering Zhejiang University Hangzhou China Beijing Teleinfo Technology Company Ltd. China Academy of Information and Communications Technology Beijing China Research Institute of Industrial Internet of Things China Academy of Information and Communications Technology Beijing China

Empirical Risk Minimization (ERM) is fragile in scenarios with insufficient labeled samples. A vanilla extension of ERM to unlabeled samples is Entropy Minimization (EntMin), which employs the soft-labels of unlabeled samples to guide their learning. However, EntMin emphasizes prediction discriminability while neglecting prediction diversity. To alleviate this issue, in this paper, we rethink the guidance information to utilize unlabeled samples. By analyzing the learning objective of ERM, we find that the guidance information for labeled samples in a specific category is the corresponding label encoding. Inspired by this finding, we propose a Label-Encoding Risk Minimization (LERM). It first estimates the label encodings through prediction means of unlabeled samples and then aligns them with their corresponding ground-truth label encodings. As a result, the LERM ensures both prediction discriminability and diversity, and it can be integrated into existing methods as a plugin. Theoretically, we analyze the relationships between LERM and ERM as well as EntMin. Empirically, we verify the superiority of the LERM under several label insufficient scenarios. The codes are available at https://***/zhangyl660/LERM. Copyright © 2024, The Authors. All rights reserved.

关键词： Forecasting

Stp-Based Verification and Synthesis of State Opacity for Logical Finite State Machines

学校读者我要写书评

暂无评论

SSRN 2022年

作者： Han, Weiwei Li, Yi Zhang, Zhipeng Xia, Chengyi School of Computer Science and Engineering Tianjin University of Technology Tianjin300384 China School of Electrical Engineering and Automation Tianjin University of Technology Tianjin300384 China School of Control Science and Engineering Tiangong University Tianjin300387 China

This paper explores the state opacity-based privacy verification and synthesis problems for finite state machines from an algebraic perspective. Firstly, the dynamics of finite state machine can be established as an algebraic expression by using the semi-tensor product (STP) of matrix. Starting from this novel representation, we propose an algebraic criterion to verify whether the privacy state is opaque to intruders. Subsequently, we investigate the opacity-enhancement synthesis problem from two aspects. On one hand, we design a feedback supervisory controller through disabling the controllable events so that the closed-loop system is opaque as far as the privacy messages are concerned;on the other hand, we investigate how to edit an observable function just by assigning some specific events in the observation channel instead of changing the transition behavior of the system. Meanwhile, we obtain that these synthesis problems can be addressed by calculating some algebraic equation system, and we further develop two algorithms to obtain the controller or function. Finally, one interesting example is illustrated to demonstrate the validity of the algebraic method. Taking together, the current results are conductive to systemically understanding the privacy enhancement issue. © 2022, The Authors. All rights reserved.

关键词： Opacity

An Efficient Genetic Algorithm for Multiple Unmanned Aerial Vehicle System Coverage Path Planning on Large-Scale Target Sets

学校读者我要写书评

暂无评论

SSRN 2024年

作者： He, Chang Ouyang, Haibin Huang, Weiqing Li, Steven Zhang, Chunliang Zhan, Zhi-Hui School of Mechanical and Electric Engineering Guangzhou University Guangzhou510006 China Guangzhou Key Laboratory of Condition Monitoring and Control of Mechanical and Electrical Equipment Guangzhou University Guangzhou510006 China Graduate School of Business and Law RMIT University Melbourne3000 Australia School of Computer Science and Engineering South China University of Technology Guangzhou China

Multiple unmanned aerial vehicle (UAV) systems have become increasingly important in various application fields, with expectations to accomplish complex tasks efficiently and with minimal resource consumption. The quality of path planning plays a crucial role in determining whether these requirements can be met. Extensive studies have been conducted on multiple unmanned aerial vehicle systems path-planning, with one prominent direction being the Genetic Algorithm (GA). However, due to the problem’s high complexity, most existing GA variants can only deal with the problem with small-scale task sets but struggle to handle the scenarios with large ones. This paper proposes an improved multi-population directed genetic algorithm (IMPDGA) for multiple drone cooperative systems path planning problems with large-scale task sets. In IMPDGA, an adaptive task volume-balanced clustering-based initialization method is designed to provide better preliminary solutions. A dominant gene evolution direction guidance strategy is employed to accelerate searching, and a local optimal escape strategy is utilized to prevent premature convergence. Additionally, two new edge assembly crossover operators are designed for the above strategies. Experiment results demonstrate that the proposed algorithm perform better than the original GA, typical GA variants, and other state-of-the-art heuristic algorithms, particularly when dealing with large-scale test sets. © 2024, The Authors. All rights reserved.

关键词： Genetic algorithms

Unsupervised Pedestrian Re-Id: Attention-Driven Framework with Clustering Optimization

学校读者我要写书评

暂无评论

SSRN 2023年

作者： Wang, Xuan Sun, Zhaojie Chehri, Abdellah Jeon, Gwanggil Song, Yongchao School of Computer and Control Engineering Yantai University Qingquan Road 30 Yantai264005 China Department of Mathematics and Computer Science Royal Military College of Canada KingstonONK7K 7B4 Canada Department of Embedded Systems Engineering Incheon National University 119 Academy-ro Yeonsu-gu Incheon22012 Korea Republic of

Unsupervised pedestrian re-identification (re-ID) is still a challenging task. Although convolutional neural networks (CNN) have had great success in pedestrian re-ID, they face the challenge of dealing with pose changes, occlusions, and lighting changes. To effectively tackle these challenges, it is imperative to prioritize the implementation of efficient sampling strategies. We propose an attention-driven framework with clustering optimization (AFC) to address the above issues. First, we introduce a new attention mechanism that enhances multi-scale spatial attention and reduces the number of trainable parameters. Then, we employed a straightforward and effective method of group sampling. In addition, we apply a clustering consensus approach to estimate pseudo-label similarity in continuous training and use temporal propagation and ensembles to improve pseudo-labels. Extensive experiments on Market-1501, DukeMTMC-reID, and MSMT17 have shown that our approach achieves significant performance gains in unsupervised pedestrian re-ID tasks, providing valuable insights for further research in this area. © 2023, The Authors. All rights reserved.

关键词： Pattern recognition

Deep Learning and Multi-Modal Fusion for Real-Time Multi-Object Tracking: Algorithms, Challenges, Datasets, and Comparative Study

学校读者我要写书评

暂无评论

SSRN 2023年

Real-time multi-object tracking (MOT) is a complex task that involves the detection and tracking of multiple objects. After the objects are detected, they are assigned markers, and their trajectories are tracked in real-time. The scientific community is intrigued bythe possibilities of utilizing MOT technology in the context of smart cities. Their primary focus lies in the domains of intelligent transportation, detection of vehicles and pedestrians, crowd surveillance, and public safety. Deep learning techniques have been developed in recent years to effectively tackle the challenges of real-time MOT tasks and enhance tracking performance. Environmental perception within smart traffic applications relies heavily on sensor data fusion. In traffic scenarios, a combination of sensors and cameras can be used to detect and track targets, as well as collect data. However, it faces challenges when it comesto detecting and tracking objects that are in motion, have complex changes in appearance, or are in crowded scenes. This paper investigates the baseline standard for real-time Multiple Object Tracking tasks. Focusing on quantitative measures, we conduct a thorough and exhaustive analysis of commonly used benchmark datasets and metrics. In this study, we also examine established embedding techniques and real-time multi-object tracking algorithms. Each strategy will be categorized and evaluated based on a set of predetermined principles. Several MOT strategies are analyzed and represented visually. Finally, this paper provides an overview of the MOT mission’s present challenges and prospective objectives. © 2023, The Authors. All rights reserved.

关键词： Smart city