检索结果-内蒙古大学图书馆

第43届中国控制会议

作者： ZHAO Pengxin Lv Qing Li Qiang Wang Jin College of Computer and Cyberspace Security Hebei Normal University College of Engineering of Hebei Normal University Hebei Normal University Hebei Provincial Key Laboratory of Information Fusion and Intelligent Control

ISBN: (数字)9789887581581

ISBN: (纸本)9798350366907

In recent years, image recognition has been commonly utilized in the field of computer vision, while the challenge of image recognition in complex environments has become a topical study. The fuzzy C-mean(FCM) is a clustering algorithm based on fuzzy set theory and achieves clustering by assigning affiliation to each data point. However, in complex environments,the FCM algorithm may fall into local optimal solutions and lead to a decrease in recognition accuracy. To solve this problem,this paper proposes an improved approach based on a genetic algorithm and simulated annealing algorithm applied in the FCM algorithm. The genetic algorithm searches and optimizes the clustering centers by simulating the biological evolution process to find more optimal parameters for the experiment. The simulated annealing algorithm performs global optimization. It generates new solutions by adding a small perturbation to avoid the clustering centers from falling into local optima. Finally, a series of image recognition experiments are conducted in complex environments. The experiment results show that the proposed approach has higher recognition accuracy and robustness in complex environments compared with the traditional FCM algorithm. The algorithm can improve the accuracy of image recognition and has good adaptability in complex environments.

关键词： Genetic algorithm simulated annealing algorithm FCM complex environment image recognition

来源：评论

学校读者我要写书评

暂无评论

Intelligent PID controller Based on Deep Reinforcement Learning

Intelligent PID Controller Based on Deep Reinforcement Learn...

引用

Robotics, control and Automation (ICRCA), International Conference on

作者： Yinhe Zhai Qiang Zhao Yinghua Han Jinkuan Wang Wenying Zeng College of Information Science and Engineering Northeastern University Shenyang China School of Control Engineering Northeastern University at Qinhuangdao Qinhuangdao China School of Computer and Communication Engineering Northeastern University at Qinhuangdao Qinhuangdao China

ISBN: (数字)9798350344721

ISBN: (纸本)9798350344738

PID control is still the most important and popular method in industrial control at present. PID control is easy to achieve and it can improve the steady-state performance and dynamic performance of the system. PID controller can be used for all objects, however, it has some problems with parameter adjustment and control effect. The proportional integral differential coefficient of PID control is fixed, and it can't change when disturbed, so the stability of the system will be affected. Moreover, PID control is prone to overshoot and can't be used in specific systems. Reinforcement learning (RL) algorithms have developed rapidly from discrete action to continuous action in recent years. It has aroused the high interest of researchers in the field of automatic control. RL control performs better in the degree of intelligence and dynamic performance, however, the steady-state performance is poor. The sensitive response of RL control will damage the actuator. In this paper, an adaptive PID controller based on deep reinforcement learning is proposed. By designing reward values, the desired control effect is described. In this way, an agent is trained to provide parameters to the PID controller in real time. It can improve the response speed of the system, suppress overshoot, and have a certain anti-disturbance ability by training the agent to achieve real-time PID parameter adjustment.

关键词： Training PI control Heuristic algorithms Industrial control Deep reinforcement learning Real-time systems Steady-state

来源：评论

学校读者我要写书评

暂无评论

Multi-branch low-light enhancement algorithm based on spatial transformation

引用

Multimedia Tools and Applications 2025年第18期84卷 19647-19667页

作者： Wang, Wenlu Sun, Ying Zou, Chunlong Tang, Dalai Fang, Zifan Tao, Bo Key Laboratory of Metallurgical Equipment and Control Technology of Ministry of Education Wuhan University of Science and Technology Wuhan430081 China School of Computer Information Management Inner Mongolia University of Finance and Economics Hohhot010051 China College of Mechanical Engineering Hubei University of Automotive Technology Shiyan442000 China Research Center for Biomimetic Robot and Intelligent Measurement and Control Wuhan University of Science and Technology Wuhan430081 China Precision Manufacturing Research Institute Wuhan University of Science and Technology Wuhan430081 China Hubei Key Laboratory of Hydroelectric Design & Maintenance China Three Gorges University Yichang443002 China

Many state-of-the-art low-light image enhancement techniques now suffer from issues like color distortion, detail blurring, and the halo effect, hindering their ability to produce visual effects. This paper presents a multi-branch low-light enhancement algorithm based on spatial transformation to significantly improve the visual impact of images under low-light settings, overcoming the limitations of existing approaches. The process involves dividing the initial low-light picture into three distinct layers. Bootstrap filtering and Contrast Limited Adaptive Histogram Equalization are used to process the main feature layer. Primary feature extraction is used to extract features for the compensation layer, which is then integrated with Contrast Limited Adaptive Histogram Equalization and MSRCR. The auxiliary layer utilizes spatial transformation to apply gamma correction to the V-channel and performs color space transformation and additional processing. Ultimately, these three layers are combined and refined to create the final improved image. Based on the experimental results, the proposed method outperforms the standard MSRCR algorithm by 52.47%, 71.58%, and 12.10%, respectively, on the assessment metrics AG, SD, and IE. LOE declines by 22.3%. The processed image presents a good visual quality, effectively enhancing the image brightness while retaining a large amount of information and improving the image details. This paper proposes an effective method to deal with the problem of low illumination image enhancement, which provides a strong reference for future research and application in related fields. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

A Learning-based Incentive Mechanism for Mobile AIGC Service in Decentralized Internet of Vehicles

A Learning-based Incentive Mechanism for Mobile AIGC Service...

引用

IEEE Conference on Vehicular Technology (VTC)

作者： Jiani Fan Minrui Xu Ziyao Liu Huanyi Ye Chaojie Gu Dusit Niyato Kwok-Yan Lam School of Computer Science and Engineering Nanyang Technological University Singapore College of Control Science and Engineering Zhejiang University People’s Republic of China

Artificial Intelligence-Generated Content (AIGC) refers to the paradigm of automated content generation utilizing AI models. Mobile AIGC services in the Internet of Vehicles (IoV) network have numerous advantages over traditional cloud-based AIGC services, including enhanced network efficiency, better reconfigurability, and stronger data security and privacy. Nonetheless, AIGC service provisioning frequently demands significant resources. Consequently, resource-constrained roadside units (RSUs) face challenges in maintaining a heterogeneous pool of AIGC services and addressing all user service requests without degrading overall performance. Therefore, in this paper, we propose a decentralized incentive mechanism for mobile AIGC service allocation, employing multi-agent deep reinforcement learning to find the balance between the supply of AIGC services on RSUs and user demand for services within the IoV context, optimizing user experience and minimizing transmission latency. Experimental results demonstrate that our approach achieves superior performance compared to other baseline models.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Assessment of Visual Fatigue Caused by Stereoscopic Disparity Based on Multimodal Measurement

SSRN

引用

SSRN 2023年

作者： Jia, Lixiu Jia, Lixin Lin, Zhong Feng, Lihang Huang, Xiaohua School of Computer Engineering Nanjing Institute of Technology Nanjing China College of Electrical Engineering and Control Science Nanjing Tech University Nanjing China

Given the prevalence of visual fatigue induced by stereoscopic display, research on the neural mechanism of visual fatigue was limited. The goal of this study was to investigate the effect of stereoscopic display with different horizontal disparities on visual fatigue based on subjective questionnaire, ophthalmological parameters, eye movement and electrocardiogram (ECG). Subjects felt more headache, uncomfortable and dry eyes after watching 70-minute 3D enhanced display with larger negative disparity than 3D normal display. Meanwhile, visual function tear film break-up time (TBUT) reduced, microsaccade mean pupil diameter (MMPD) increased and heart rate variability (HRV) changed more significantly after watching 3D enhanced display. Overall, these results demonstrated that 3D enhanced display had a better sense of immersion, but it induced more visual fatigue significantly than 3D normal display. © 2023, The Authors. All rights reserved.

关键词： Electrocardiography

来源：评论

学校读者我要写书评

暂无评论

Fractional Order Sliding Mode control with Fuzzy Parameter Tuning Applied to a SCARA Robot

Fractional Order Sliding Mode Control with Fuzzy Parameter T...

引用

Mechatronic Systems Supervision (IW_MSS), IEEE International Workshop on

作者： Aicha Znidi Fatma Abdelhedi Ahmed Said Nouri Nabil Derbel Lab: Numerical Control of Industrial Processes National Engineering School of Gabes University of Gabes Tunisia Department of Electrical and Computer Engineering College of Engineering King Abdulaziz University Jeddah Saudi Arabia Control & Energy Management Laboratory University of Sfax National Engineering School of Sfax Tunisia

ISBN: (数字)9798350378382

ISBN: (纸本)9798350378399

This paper presents the application of Fractional Order Sliding Mode control (FO-SMC) in order to achieve a robust motion trajectory regulation in dynamic robot systems. The proposed control strategy benefits of both advantages of fractional calculus and sliding mode control to upgrade the robustness and the performances of the controller. Moreover, the paper proposes to use a fuzzy tuning of sliding mode controller in order to satisfy the capabilities of the actuators. The stability of the closed loop system has been demonstrated using the Lyapunov theory. Simulation results demonstrate the efficiency of the proposed method achieving precise trajectory tracking and robust performances, it has been shown that fuzzy tuning of the sliding function parameters significantly reduces control variable peaks, making the control more suitable for actuator application.

关键词： Asymptotic stability Actuators Trajectory tracking Simulation Manipulators Stability analysis Closed loop systems Tuning Sliding mode control Fractional calculus

来源：评论

学校读者我要写书评

暂无评论

Applying Transformer-Based Neural Networks to Corpus Linguistic Data for Predictive Text Generation in Multilingual Environments 6

Applying Transformer-Based Neural Networks to Corpus Linguis...

引用

6th IEEE International Conference on Emerging Smart Computing and Informatics, ESCI 2024

作者： Taufikin Orosoo, Myagmarsuren Rengarajan, Manikandan Fatma, Gulnaz Yuldashev, Dilyorjon Raj, I Infant IAIN Kudus Indonesia Mongolian National University of Education Mongolia Vel Tech Rangarajan Dr. Sagunthala R and D Institute of Science and Technology Tamil Nadu Chennai India University College Al Ardah Jazan University Jazan Saudi Arabia Turin Polytechnic University in Tashkent Automatic Control and Computer Engineering Department Tashkent Uzbekistan K. Ramakrishnan College of Engineering Department of CSE Tamil Nadu Trichy India

ISBN: (纸本)9798350306613

In the subject of corpus linguistics, this study explores the use of transformer-based neural networks to the problems of predictive text production in multilingual contexts. The Transformer architecture presents a viable framework for natural language processing problems because of its reputation for capturing complex patterns and long-range dependencies in sequential data. The research employs a two-phase approach, whereby the Transformer is first pre-Trained using a variety of multilingual corpora. This helps the model learn representations that are independent of language, which makes it easier for it to adapt to various linguistic circumstances. The model is then fine-Tuned using language-specific datasets to improve its ability to produce linguistically nuanced and contextually appropriate text. The study looks into how different model architectures, training methods, and hyperparameters affect how well the suggested multilingual predictive text generation system performs. Evaluation measures are used to evaluate the model's performance in several languages, including linguistic diversity measurements, BLEU scores, and perplexity. The overall accuracy, calculated across all languages, stands at 80.3%, indicating the model's robust cross-lingual generalization and its effectiveness in capturing diverse linguistic nuances The efficacy and generalization of Transformer-based models are improved across a wide range of languages to handle linguistic diversity. © 2024 IEEE.

关键词： Corpus Linguistics Cultural Context Integration Multilingual Text Generation Neural Language Models Transformer-Based Neural Networks

来源：评论

学校读者我要写书评

暂无评论

LEARNINGS FROM IMPLEMENTATION OF A DETUMBLING ALGORITHM FOR A NANOSATELLITE 49

LEARNINGS FROM IMPLEMENTATION OF A DETUMBLING ALGORITHM FOR ...

引用

49th Student Conference 2021 at the 72nd International Astronautical Congress, IAC 2021

作者： Hire, Shivam Bhakare, Onkar Mahajan, Sushil Karaguppi, Atharva Thorat, Shubham Patil, Rahul Durgule, Manali Shaikh, Juber Pardeshi, Bhaskar Patel, Kishan Bandukwala, Idris Rathod, Akash Deodhar, Akshay Gupta, Aman Department of Computer Engineering and Information Technology College of Engineering Pune India Department of Electrical Engineering College of Engineering Pune India Department of Electronics Engineering College of Engineering Pune India Department of Mechanical Engineering College of Engineering Pune India Department of Instrumentation and Control College of Engineering Pune India

ISBN: (纸本)9781713843184

Nanosatellites, on ejection from the launch vehicle, tumble with random angular velocities and have to be stabilized before they can commence their normal working. Due to the restriction on the mass and size of the nanosatellite, the use of exhaustive massive thrusters is not feasible. Hence, nanosatellites must use alternative methods to stabilize themselves. For the COEPSAT-2 nanosatellite, the B-Dot algorithm will be used to damp the angular velocities along all 3 axes. B-Dot uses the torque generated by the interaction between the torque rods and the earth's magnetic field to stabilize the satellite. This paper presents the implementation, testing procedure, and experimental results obtained during the testing of the B-dot algorithm. The B-Dot algorithm requires control of magnetic dipole moment, which is achieved by controlling the current through the magnetorquer coils. A low-power controlled current source was developed using linear regulator topology. The detailed design procedure and testing results of the current control circuit are presented in this paper. A design of an alternative switch mode topology is also discussed in the paper. For on-ground testing, an air-bearing was used for levitating the test setup, allowing it to rotate freely about one axis. This enabled us to evaluate the damping effect of the B-Dot algorithm on angular velocity in space-like conditions. An exponential decay in angular velocity was observed in these experiments. We present the angular velocity data acquired in these experiments and the effects that various factors like magnetic field strength, maximum magnetorquer current, gain values, initial velocity, and sensor calibration have on stabilization time. Copyright © 2021 by the International Astronautical Federation (IAF). All rights reserved.

关键词： Stabilization

来源：评论

学校读者我要写书评

暂无评论

Observer-based multivariable fixed-time formation control of mobile robots

引用

Journal of Systems engineering and Electronics 2020年第2期31卷 403-414页

作者： LI Yandong ZHU Ling and GUO Yuan College of Computer and Control Engineering Qiqihar UniversityQiqihar 161006China School of Mechanical and Electronic Engineering Qiqihar UniversityQiqihar 161006China

This paper proposes a multivariable fixed-time leaderfollower formation control method for a group of nonholonomic mobile robots, which has the ability to estimate multiple uncertainties. Firstly, based on the state space model of the leader-follower formation, a multivariable fixed-time formation kinematics controller is designed. Secondly, to overcome uncertainties existing in the nonholonomic mobile robot system, such as load change,friction, external disturbance, a multivariable fixed-time torque controller based on the fixed-time disturbance observer at the dynamic level is designed. The designed torque controller is cascaded with the formation controller and finally realizes accurate estimation of the uncertain part of the system, the follower tracking of reference velocity and the desired formation of the leader and the follower in a fixed-time. The fixed-time upper bound is completely determined by the controller parameters, which is independent of the initial state of the system. The multivariable fixed-time control theory and the Lyapunov method are adopted to ensure the system ***, the effectiveness of the proposed algorithm is verified by the experimental simulation.

关键词： multivariable fixed-time control formation control uncertainty fixed time observer nonholonomic mobile robot

来源：评论

学校读者我要写书评

暂无评论

A Coding-Enhanced Jamming Approach for Secure Semantic Communication over Wiretap Channels

arXiv

引用

arXiv 2025年

作者： Chen, Weixuan Q., Yang S., Shao Z., Shi J., Chen X., Shen College of Information Science and Electronic Engineering Zhejiang University Hangzhou China School of Cyber Science and Engineering Shanghai Jiao Tong University Shanghai China College of Control Science and Engineering Zhejiang University Hangzhou China Department of Electrical and Computer Engineering University of Waterloo Waterloo Canada

As semantic communication (SemCom) gains increasing attention as a novel communication paradigm, ensuring the security of transmitted semantic information over open wireless channels becomes crucial. Existing secure SemCom solutions often lack explicit control over security. To address this, we propose a coding-enhanced jamming approach for secure SemCom over wiretap channels. This approach integrates deep joint source and channel coding (DeepJSCC) with neural network-based digital modulation, enabling controlled jamming through two-layer superposition coding. The outer constellation sequence encodes the source image, while the inner constellation sequence, derived from a secret image, acts as the jamming signal. By minimizing the mutual information between the outer and inner constellation sequences, the jamming effect is enhanced. The jamming signal is superposed on the outer constellation sequence, preventing the eavesdropper from recovering the source image. The power allocation coefficient (PAC) in the superposition coding can be adjusted to control system security. Experiments show that our approach matches existing methods in security while significantly improving reconstruction performance across varying channel signal-to-noise ratios (SNRs) and compression ratios. Copyright © 2025, The Authors. All rights reserved.

关键词： Digital communication systems

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：