检索结果-内蒙古大学图书馆

2024 IEEE International Conference on Real-Time Computing and Robotics, RCAR 2024

作者： Yu, Guanfeng Zhang, Han Wang, Zehao Wang, Jingchuan Shanghai Jiao Tong University Institute of Medical Robotics Department of Automation China Ministry of Education of China Key Laboratory of System Control and Information Processing China Shanghai Engineering Research Center of Intelligent Control and Management Shanghai200240 China

ISBN: (纸本)9798350372601

The paper introduces a Koopman bilinear model predictive control (KBMPC) as a local planner for smart wheelchair systems. This approach leverages the Koopman Operator to streamline the nonlinear terms in MPC, aiming to achieve superior real-time performance. In particular, we employ the Extended Dynamic Mode Decomposition (EDMD) method to approximate the infinite dimensional Koopman operator. To tackle the challenge of designing lifting functions, we systematically utilize the derivatives of known nonlinear structures to methodically derive lifting functions. This systematic approach facilitates the identification of a precise bilinear model. Next, we formulate the corresponding Koopman bilinear MPC problem. Finally, an iterative algorithm is employed to efficiently solve the KBMPC problem. This method linearizes the bilinear model around estimations and iteratively solves the resulting quadratic programming problem to refine estimations. This allows KBMPC to function as a real-time local planner for smart wheelchairs. The prediction of nonlinear terms shows that the identified Koopman bilinear model increases the prediction accuracy. Furthermore, experiments demonstrate the advantages of the designed KBMPC local planner, highlighting its superior real-time performance and predictive capabilities. © 2024 IEEE.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

Self-Supervised Learning with Min-Max Distance Approach for Few-Shot Steel Surface Defects Recognition 43

Self-Supervised Learning with Min-Max Distance Approach for ...

引用

43rd Chinese control Conference, CCC 2024

作者： Wang, Tianlei Chen, Xinyao Li, Zeliang Zhou, Zhencai Zhou, Zhiheng Wuyi University School of Electronic and Information Engineering Jiangmen China Jiangmen Key Laboratory of Kejie Semiconductor Bonding Technology and Control System Jiangmen China Wuyi University School of Mechanical and Automation Engineering Jiangmen China Guangdong Province Kejie Machinery Automation Company Technical Department Jiangmen China

ISBN: (纸本)9789887581581

Surface defects in steel are crucial for quality inspection in the manufacturing of industrial metal products. While deep learning methods have been extensively studied in this field, the lack of large-scale annotated defect samples imposes significant labeling pressure, making training fully supervised deep models challenging. In response to this issue, we employ self-supervised learning for few-shot recognition, effectively learning representations from limited unlabeled data and achieving defect recognition with a small amount of labeled samples. In particular, we construct Other encoder with multi-view information and Base encoder with single view information based on the distillation self-supervised learning framework through the multi-crops mechanism, and the Other encoders and Base encoders perform information share. Our main contribution is that, based on this framework, we propose a new regularization method to solve the problem of instances deviating from the center of self-supervised clustering by minimizing the maximum instance distance (MMD). In testing, fine-tuning the MMD pre-trained model with a small number of labeled data achieves superior recognition performance on the steel surface defect dataset NEU, significantly outperforming other testing methods. © 2024 Technical Committee on control Theory, Chinese Association of automation.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Persistent Coverage in Non-Convex Environments with Heterogeneous Robotic Networks: Constraint-based Approach ⋆

引用

IFAC-PapersOnLine 2023年第2期56卷 10708-10714页

作者： J.G. Martin T. Hatanaka J.M. Maestre E.F. Camacho Department of Systems and Automation Engineering University of Seville Spain Department of Systems and Control Engineering Tokyo Institute of Technology

In this paper, we extend a constraint-based coverage control for robotic sensor networks based on control barrier functions (CBFs) for environments with known obstacles and different types of unmanned vehicles (UV). To this end, we use a sensing function that considers the vertices of the obstacles to compute the route and the distance regarding the UVs. This way, obstacle avoidance becomes intrinsic to the CBF ensuring the coverage performance. Moreover, we consider different obstacles and speeds for each type of UV. Finally, the proposed algorithm is illustrated with a heterogeneous fleet of UVs and obstacles in a simulated thermosolar power plant.

关键词： Coordination of multiple vehicle systems Graph-based methods for networked systems Networked robotic systems control barrier function (CBF) coverage control obstacles non-convex area

来源：评论

学校读者我要写书评

暂无评论

Lyapunov control Method Based Optimal control of Alpha-Type Stirling Motor 4

Lyapunov Control Method Based Optimal Control of Alpha-Type ...

引用

4th International Congress on Human-Computer Interaction, Optimization and Robotic Applications, HORA 2022

作者： Azanpa, Mahmut Vural, Berkem Tas, Ali Ihsan Ozden, Ali Burak Iscan, Mehmet Yildiz Technical University Mechatronics Engineering Department Istanbul Turkey Yildiz Technical University Control and Automation Engineering Department Istanbul Turkey

ISBN: (纸本)9781665468350

Nowadays, the optimally developed system design has been important with the improvement of the technology. The mathematical model approach with the proper controller design provides a more efficient system with its outputs, as well. One of the complex structures is the alpha type Stirling motor in which it is used to control two independent pistons for gaining more power and but makes controlling its dynamics difficult due to its complex structure dynamics in terms of heat and mechanics interactions. In addition to these difficulties, it must be required to design optimal control via system dynamics perspectives because of the characteristics of four different phases. Due to these reasons, in this study, nonlinear Lyapunov-based optimal control for alpha type stirling engine was performed by combining ideal thermodynamic and mechanic equations. The suggested controller was tested under the different reference motor velocities by calculating optimal heat energy consumption via the control rule. After designing two cascade controllers, both mechanical and heat system controls were achieved to optimize heat energy consumption for different motor velocities. It was proved that the proposed method attained high performance for reference tracking of alpha Stirling motor velocities by interacting heat-mechanical system dynamics modeling via Lyapunov control theory. © 2022 IEEE.

关键词： control theory

来源：评论

学校读者我要写书评

暂无评论

Fault Diagnosis Based on the Updating Strategy of Interval-Valued Belief Structures

引用

Chinese Journal of Electronics 2025年第4期23卷 753-760页

作者： Xiaobin Xu Ping Liu Yanbo Sun Chenglin Wen Institute of Systems Science and Control Engineering School of Automation Hangzhou Dianzi University Hangzhou China Department of Control Science and Engineering School of Astronautics Harbin Institute of Technology Harbin China

This paper presents the dynamic method for fault diagnosis based on the updating of Interval-valued belief structures (IBSs). The classical Jeffrey's updating rule and the linear updating rule are extended to the framework of IBSs. Both of them are recursively used to generate global diagnosis evidence with the form of Interval basic belief assignment (IBBA) by updating the previous evidence with the incoming evidence. The diagnosis decision can be made by global diagnosis evidence. In the process of evidence updating, the similarity factors of evidence are used to determine switching between the extended Jeffrey's and linear updating rules, and to calculate the linear combination weights. The diagnosis examples of machine rotor show that the proposed method can provide more reliable and accurate results than the diagnosis methods based on Dempster-Shafer evidence theory.

关键词： Fault diagnosis Accuracy Evidence theory Rotors Switches Reliability theory Real-time systems

来源：评论

学校读者我要写书评

暂无评论

LI-DPS: A Long Sequence Dual Prediction Scheme Based on Informer for Efficient High-Frequency Data Transmission 50

LI-DPS: A Long Sequence Dual Prediction Scheme Based on Info...

引用

50th Annual Conference of the IEEE Industrial Electronics Society, IECON 2024

作者： Wu, Yu Yang, Bo Zhu, Dafeng Chen, Cailian Department of Automation Shanghai Jiao Tong University Shanghai200240 China Key Laboratory of System Control and Information Processing Ministry of Education of China China Shanghai Engineering Research Center of Industrial Intelligent Control and Management Shanghai200240 China

ISBN: (纸本)9781665464543

High-frequency time-series data such as vibration signals will consume a lot of communication resources and require very high network bandwidth. Reducing the amount of data transmission while ensuring its availability is particularly important and challenging in high-frequency scenarios. Dual prediction scheme (DPS) can significantly reduce the amount of data transmission while ensuring data accuracy. However, traditional DPS cannot be applied to high-frequency scenarios due to the limitation of the mechanism design and model inference speed. We propose a novel long sequence DPS to efficiently reduce the volume of high-frequency data transmission in real time. To overcome the accuracy degradation due to gradient vanishing in long sequence prediction, we also propose an attention-based prediction model. Furthermore, we propose Informer to reduce the model’s computational complexity and enable fast inference and deployment on resource-constrained edge gateways. Experiments based on the real-world dataset show that the proposed scheme can efficiently cope with high-frequency data, reduce its transmission volume, and ensure data accuracy. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Real Time Discrete control Algorithm for Gimbal System Design in Mini Unmanned Aerial Vehicles 4

Real Time Discrete Control Algorithm for Gimbal System Desig...

引用

4th International Congress on Human-Computer Interaction, Optimization and Robotic Applications, HORA 2022

作者： Tas, Ali Ihsan Iscan, Mehmet Ozden, Ali Burak Cinar, Emre Vural, Berkem Yilmaz, Cuneyt Yildiz Technical University Mechatronics Engineering Department Istanbul Turkey Yildiz Technical University Control and Automation Engineering Department Istanbul Turkey

ISBN: (纸本)9781665468350

Nowadays, gimbal systems have been a very critical role to obtain some important information related to live images photos and target detection in unmanned aerial vehicles. The most basic performance criterion of gimbal systems depends on its stabilization in which the desired reference tracking must be accomplished by designing complex controller structure. For this purpose, a discrete controller design is presented to provide reduced cost, high-performance, 3-axis gimbal systems in real-time operation. Firstly, the mathematical model of the system was obtained and 3D - computer aided design was made in order to calculate the system parameters like inertia, damping factor, etc. After that forward and inverse kinematic equations for the gimbal system were obtained in order to calculate the reference degree in each axis. The discrete controller was derived from energy-based computed torque proportional controller which was based on the Lyapunov Stability Theorem in discrete notation. The reference tracking results at 10 milliseconds sampling time are as follows;0 overshoot for 3 axes with lack of steady-state errors, and the time constants are 0.1575, 0.1610, and 0.27 second which are acceptable for roll, pitch, and yaw axes, respectively. In addition, the discrete-time performance of the proposed controller was compared with a PID controller designed in continuous time, and it was shown that the continuous-time design did not work efficiently as well as discrete time one. © 2022 IEEE.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Heat Transport Modeling And Model-Based Temperature control Of A Residential Building, A Case Study 24

Heat Transport Modeling And Model-Based Temperature Control ...

引用

24th International Carpathian control Conference, ICCC 2023

作者： Gal, Krisztian Szilveszter Nimrod Kutasi, Denes Automation Engineer Thermo Control Srl. Sfantu Gheorghe Romania Sapientia Hungarian University of Transylvania Department of Electrical Engineering Tirgu Mures Romania

ISBN: (纸本)9798350310221

The paper proposes a case study on modeling, model validation, model-based temperature control of a building to demonstrate the effectiveness of a modern control technique versus a classical control used in a residential building. Energy efficiency is a key topic today, thus it is important to study the methods of temperature control in buildings to achieve the desired control with minimum energy usage. The paper presents the heat transport modeling using the R-C model technique, the model validation (parameter tuning) based on long-term measurements, classical control and model-based predictive (MPC) distributed temperature control in the rooms of the building and the comparison of the energy consumption for each case. Moreover, there are two MPC algorithms discussed, with and without external temperature forecast as disturbance variable. The simulation results and real-time measurements are presented. © 2023 IEEE.

关键词： Model predictive control

来源：评论

学校读者我要写书评

暂无评论

A Pre-Distillation Strategy for Object Detection Task 9

A Pre-Distillation Strategy for Object Detection Task

引用

9th International Conference on control, Decision and Information Technologies, CoDIT 2023

作者： Lin, Yu-Chen Chen, Wen-Hui Chen, Hung-Cheng Wang, Chia-Hung Feng Chia University Department of Automatic Control Engineering Taichung Taiwan Graduate Institute of Automation Technology National Taipei University of Technology Taipei Taiwan

ISBN: (纸本)9798350311402

In this paper, we propose a novel model compression method based on knowledge distillation, that is called Pre-Distillation, for object detection task. First, the full model is 'squashed' (by the instructor) to generate key information (a.k.a. attention mechanism), and it is also referred to as dataset condensation;it is then provided for lightweight model learning (for the student). Students can study and preview the course through lightweight models with key information that is compressed by the instructor. Additionally, students can also be able to improve their efficiency in learning and acquire integrative understanding of the subject through supplementary materials in which the they will be observing hidden information within the ground truth. This paper is verified and quantified by the public databases of Pascal VOC. The results show that the previewing distillation strategy developed in this paper has higher accuracy. The accuracy of ResNet-101 is 74.02%, which is 2.28% higher than that of the un-distilled 71.74%. The method in this paper is an algorithm-based performance improvement method. Our proposed scheme is a high-efficiency distillation strategy that does not need to change the network architecture and also can avoid the occurrence of model overfitting and speed up model training. © 2023 IEEE.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Computationally Efficient Core Losses Calculation and Thermal Evaluation Using the VBR-Based Finite Element Model of Induction Motors

Computationally Efficient Core Losses Calculation and Therma...

引用

2023 IEEE Energy Conversion Congress and Exposition, ECCE 2023

作者： Valipoor, Zahra Koti, Hossein Nejadi Demerdash, Nabeel A. O. Marquette University Electrical and Computer Engineering Department Milwaukee United States Rockwell Automation Inc. Motion Control Team Mequon United States

ISBN: (纸本)9798350316445

Accuracy of the simulation results obtained from the FEA analysis of electric motors is highly dependent on the quality of the motor performance metrics such as motor steady state temperatures and losses. In addition such FEA analysis is typically computationally burdensome and demanding. Improving the computational burden and accuracy of core losses calculation and thermal evaluation in Time-Stepping Finite Element (TS-FE) models of induction motors is addressed in this paper. The Virtual Blocked Rotor (VBR)-based Time-Stepping Finite Element (TS-FE) technique is utilized to mitigate the numerical transient response of the induction motor FE model. The motor core losses have been calculated through the harmonic based core loss calculation that assumes variable loss coefficients as a function of magnetic flux density. In addition, the motor estimated losses, including the calculated core losses are fed to a thermal network model to estimate the temperature of different motor components. © 2023 IEEE.

关键词： Core Losses Induction Motor Temperature Thermal Calibration VBR-based FEA

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：