检索结果-内蒙古大学图书馆

SPIE 12th International Symposium on Multispectral Image Processing and Pattern Recognition, MIPPR 2023

作者： Mu, HaoDong Li, Haipeng Chen, Zhong Huazhong University of Science and Technology 1037 Luoyu Road HuBei WuHan430074 China Beijing Aerospace Automatic Control Institute Bejing100854 China

ISBN: (数字)9781510675001

ISBN: (纸本)9781510674998

As remote sensing technology continues to advance, the accuracy and quantity of remote sensing images have significantly improved. The generation of a vast amount of available data has facilitated the widespread application of various deep learning methods in the field of remote sensing data processing, such as object detection, semantic segmentation, and change detection. In the aforementioned tasks, Change detection is used to identify alterations occurring on the Earth's surface by utilizing remote sensing (RS) data. In recent years, deep learning based methods have exhibited significantly superior performance compared to traditional change detection techniques. The fundamental strategy enabling these advancements involves extracting appropriate deep learning features from input remote sensing images through various backbone networks such as VGG, ResNet, DenseNet etc. Nevertheless, the features extracted by the aforementioned backbone networks may not fully cater to the specific requirements of remote sensing image change detection tasks. Consequently, our goal is to explore the influence of features extracted by different backbone networks on change detection tasks and introduce a specialized backbone network tailored for change detection. This endeavor aims to produce features that are better suited for the of change detection. The experimental results indicate that our specifically designed feature extraction network for remote sensing image change detection outperforms traditional networks in extracting task-specific features. These features are better suited for subsequent decoder modules, enhancing the generation of image-based change detection results. At the same time, we found that when using general backbone networks for change detection, ResNet achieves the highest metric accuracy, while DenseNet has the lowest memory usage and the fastest training and testing speed. Depending on the specific task, we can choose the appropriate backbone network as need

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Integrated Guidance and control System Design for Hypersonic Vehicle Based on Backstepping High-Order Sliding Mode Approach 42

Integrated Guidance and Control System Design for Hypersonic...

引用

42nd Chinese control Conference, CCC 2023

作者： Zhuo, Xiaobin Zhang, Xiangyin Yu, Yue Beijing University of Technology Faculty of Information Technology Beijing100124 China Beijing Aerospace Automatic Control Institute Beijing100854 China

ISBN: (纸本)9789887581543

A six-degree-of-freedom (6DOF) control scheme is proposed for the glide phase of hypersonic vehicle. Based on hypersonic vehicle motion and engagement model, the 3D integrated guidance and control (IGC) model is constructed. The backstepping-high order sliding mode (BHOSM) control scheme is proposed based on backstepping method and the high-order sliding mode (HOSM) method. The backstepping controller is designed to solve the problem of under-actuation in the IGC model. The HOSM controller is designed to overcome the uncertainty, strong coupling and nonlinear problems of IGC system. According to Lyapunov's theorem, the states of the closed loop system are proved to be uniform bounded. Finally, the effectiveness and reliability of the proposed method is verified and studied by simulation experiments. © 2023 Technical Committee on control Theory, Chinese Association of Automation.

关键词： Backstepping

来源：评论

学校读者我要写书评

暂无评论

Multi-Agent Hierarchical Decision Optimization Method Based on Curriculum Learning 9

Multi-Agent Hierarchical Decision Optimization Method Based ...

引用

9th International Conference on Big Data and Information Analytics, BigDIA 2023

作者： Liang, Wenqian Huang, Chengqi Zhang, Yong Zhang, Zhuxin Yang, Xiaolong Zhang, Pengcheng National University of Defense Technology Laboratory for Big Data and Decision Changsha China North Automatic Control Technology Institute Taiyuan China

ISBN: (纸本)9798350330076

The huge state space and varying action space will make the decision-making be a grand challenge for reinforcement learning. To address this problem, we propose a multi-agent hierarchical decision optimization algorithm, which utilizes macro-action abstraction based on expert experience knowledge and hierarchical abstraction based on hierarchical reinforcement learning theory to decompose the combat task into different hierarchical sub-task modules. The alternating update method is used to learn and train the network. Meanwhile, combined with the curriculum learning method, the training process of the neural network model can be improved. The algorithm is validated in the full-length game of StarCraft II. When playing against the built-in AI of level 7 difficulty, it achieved a win rate of over 93%. The experimental results show that it can effectively improve the collaborative ability of the decision-making model when the state space and the optional action space are huge and the decision-making process is complex. © 2023 IEEE.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Coupled Output Synchronization for Leader-Follower Vehicle Platooning 8

Coupled Output Synchronization for Leader-Follower Vehicle P...

引用

8th International Conference on Robotics and Automation Engineering, ICRAE 2023

作者： Sinha, Pallavi Dutta, Maitreyee Srikant, Sukumar Systems and Control Engineering Indian Institute of Technology Bombay Powai Mumbai400076 India

ISBN: (纸本)9798350327656

A novel framework in multi-agent system networks is introduced and analyzed in this article, namely coupled output synchronization over directed graph topologies. Under assumptions of a rooted spanning tree, we design a distributed control law for synchronization of Laplacian coupled outputs of double integrator dynamics. The analysis of the closed-loop systems communicating over digraphs is carried out via Lyapunov methods. The vehicle platooning problem is demonstrated as a direct application of our results on coupled output synchronization. The control is distributed and independent of the knowledge of eigenvalues of the graph Laplacian. Simulations verify the performance of the controller with and without disturbance. © 2023 IEEE.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Reduced order model with non-linear parameter optimization for hydrogen combustion engine control

Reduced order model with non-linear parameter optimization f...

引用

2023 IEEE Conference on control technology and Applications, CCTA 2023

作者： Kluge, Kevin Pehlivanlar, Benjamin Wenzel, Henrik Gunther, Marco Pischinger, Stefan Abel, Dirk Rwth Aachen University Institute of Automatic Control Germany Rwth Aachen University Chair of Thermodynamics of Mobile Energy Conversion Systems Germany

ISBN: (纸本)9798350335446

Using hydrogen as a fuel in spark ignition (SI) internal combustion engines offers an environmentally friendly alternative to fossil fuels, while retaining many benefits of conventional powertrains. Even though the hydrogen SI engine can rely on mature mechanical parts from conventional gasoline engines, the control software has to be adapted to the change in fuel. This includes a rapid switching between lean and stochiometric operation while ensuring engine protection and very low pollutant emissions. A model predictive controller (MPC) offers the possibility to systematically consider opposing control objectives and constraints for transient operation while reducing the development and calibration effort. As a basis for the real-time capable MPC, a reduced order physical model with empirical components is developed to describe the processes in the combustion chamber. Taking single cylinder engine testbench measurements into account from early stages of the engine development, some of the model's parameters are optimized to increase prediction accuracy. The resulting non-linear parameter optimization reduces the mean square errors of the main outputs relative air-fuel-ratio and indicated mean effective pressure from 0.1475 and 0.5426 bar2 to 0.0218 and 0.1020 bar2 within a wide operating area. © 2023 IEEE.

关键词： Hydrogen

来源：评论

学校读者我要写书评

暂无评论

Multi-objective robust controller synthesis with integral quadratic constraints in discrete-time

arXiv

引用

arXiv 2025年

作者： Schwenkel, Lukas Köhler, Johannes Müller, Matthias A. Scherer, Carsten W. Allgöwer, Frank University of Stuttgart Institute for Systems Theory and Automatic Control Germany ETH Zürich Institute for Dynamical Systems and Control Switzerland Leibniz University Hannover Institute for Automatic Control Germany University of Stuttgart Department of Mathematics Germany

This article presents a novel framework for the robust controller synthesis problem in discrete-time systems using dynamic Integral Quadratic Constraints (IQCs). We present an algorithm to minimize closed-loop performance measures such as the H∞-norm, the energy-to-peak gain, the peak-to-peak gain, or a multi-objective mix thereof. While IQCs provide a powerful tool for modeling structured uncertainties and nonlinearities, existing synthesis methods are limited to the H∞-norm, continuous-time systems, or special system structures. By minimizing the energy-to-peak and peak-to-peak gain, the proposed synthesis can be utilized to bound the peak of the output, which is crucial in many applications requiring robust constraint satisfaction, input-to-state stability, reachability analysis, or other pointwise-in-time bounds. Numerical examples demonstrate the robustness and performance of the controllers synthesized with the proposed algorithm. Copyright © 2025, The Authors. All rights reserved.

关键词： Robust control

来源：评论

学校读者我要写书评

暂无评论

Path-Complete Barrier Functions for Safety of Switched Linear systems 63

Path-Complete Barrier Functions for Safety of Switched Linea...

引用

63rd IEEE Conference on Decision and control, CDC 2024

作者： Anand, Mahathi Jungers, Raphael Zamani, Majid Allgower, Frank Institute for Systems Theory and Automatic Control University of Stuttgart Germany Université Catholique de Louvain Department of Mathematical Engineering Belgium University of Colorado Boulder United States

ISBN: (纸本)9798350316339

In this paper, we address the safety verification problem of switched linear dynamical systems under arbitrary switching via barrier functions. Our approach is based on a notion of path-complete barrier functions, which utilizes a collection of barrier functions associated with a directed labeled graph that can encode all the possible switching sequences. We show that path-complete barrier functions effectively generalize notions of common and multiple barrier functions studied in existing literature, and can potentially provide less conservative conditions for safety verification. We demonstrate that, for switched linear systems, the inequalities imposed via path-complete barrier functions can be easily encoded into simple linear matrix inequalities under some assumptions on the regions of interest and appropriately chosen templates for the barrier functions. We also study the relationship between path-complete barrier functions and common barrier functions, and show that for any path-complete graph with an admissible path-complete barrier function, one can derive a suitable (possibly non-smooth) common barrier function by utilizing the path-complete barrier function. Finally, we utilize several examples to illustrate the effectiveness of our approach, and briefly discuss the challenges that lay foundations for future research. © 2024 IEEE.

关键词： Directed graphs

来源：评论

学校读者我要写书评

暂无评论

Full-Parameter Fine-Tuning Method of LLMs for Sports Injury Prevention and Treatment

引用

International Journal of Mobile Computing and Multimedia Communications 2025年第1期16卷

作者： Zhu, Xinli Gao, Zhiqiang Wang, Xu An North Automatic Control Technology Institute China Engineering University of PAP China

Fine-tuning large language models (LLMs) for sports injury prevention and treatment in resource-constrained environments poses significant challenges due to memory demands and growing size of data. This paper proposes an efficient full-parameter fine-tuning approach based on Gradient Low-Rank Projection (GaLore) to reduce memory usage. Further, a data augmentation strategy for sports injury prevention and treatment is utilized to finetune a question-and-answer (Q&A) model with 0.5B parameter on consumer GPUs with 24GB memory. Experiment results show that the proposed method enhanced by GaLore is superior to SOTA methods such as low-rank adaptation (LoRA) in terms of convergence accuracy, training time, memory consumption, and indicators of BLEU-4 and ROUGE-2. Meanwhile, the empirical effect of injury prevention Q&A cases indicate that Qwen2-0.5B-Instruct trained by the proposed method have obvious advantages in professional knowledge understanding and overcoming hallucinations. © 2025 IGI Global. All rights reserved.

关键词： Sports medicine

来源：评论

学校读者我要写书评

暂无评论

On the Regret of Recursive Methods for Discrete-Time Adaptive control with Matched Uncertainty

On the Regret of Recursive Methods for Discrete-Time Adaptiv...

引用

IEEE Conference on Decision and control

作者： Aren Karapetyan Efe C. Balta Anastasios Tsiamis Andrea Iannelli John Lygeros Automatic Control Laboratory Swiss Federal Institute of Technology (ETH Zürich) Zürich Switzerland Control and Automation Group inspire AG Zürich Switzerland Automatic Control Laboratory Institute for Systems Theory and Automatic Control University of Stuttgart Stuttgart Germany

ISBN: (数字)9798350316339

ISBN: (纸本)9798350316346

Continuous-time adaptive controllers for systems with a matched uncertainty often comprise an online parameter estimator and a corresponding parameterized controller to cancel the uncertainty. However, such methods are often impossible to implement directly, as they depend on an unobserved estimation error. We consider the equivalent discrete-time setting with a causal information structure, and propose a novel, online proximal point method-based adaptive controller, that under a sufficient excitation (SE) condition is asymptotically stable and achieves finite regret, scaling only with the time required to fulfill the SE. We show the same also for the widely-used recursive least squares with exponential forgetting controller under a stronger persistence of excitation condition.

关键词： Estimation error Uncertainty Costs Noise Minimization control systems Adaptive control

来源：评论

学校读者我要写书评

暂无评论

Optimization Design and Dynamic Compensation Strategy of Inverter Parallel Virtual Impedance Based on Residual Generator 7

Optimization Design and Dynamic Compensation Strategy of Inv...

引用

7th IEEE Conference on Energy Internet and Energy System Integration, EI2 2023

作者： Hu, Changbin Huang, Haiyang Luo, Shanna Wang, Huiru Li, Xuecheng Lu, Heng North China University of Technology School of Electrical and Control Engineering Beijing China Beijing Kedong Electric Power Control System Co. Ltd Beijing China University of Duisburg-Essen Department of Automatic Control and Complex Systems Germany

ISBN: (纸本)9798350345094

To solve the power imbalance problem caused by the difference between the equivalent impedance of the inverter and the line impedance, and the power quality problem caused by the disturbance of load mutation, unbalanced load and nonlinear load, this paper proposes a closed-loop feedback dynamic compensation strategy based on the equivalent virtual impedance of the residual generator. Firstly, based on the parallel topology of inverters, a quasi-steady-state model is established on the dq axis. Based on the model, the power imbalance caused by the difference of output impedance and line impedance of different inverters and the power quality problems caused by load switching, unbalanced load, nonlinear load and other disturbances in parallel inverters are ***, a residual generator is established based on the inverter model. A virtual impedance equivalent controller Qvir(S) based on the residual generator is designed by the Luenberger observer principle. The disturbance information is fed back in real time through the feedback structure, and then the virtual impedance is adjusted to realize the power sharing of the inverter parallel system. Then, based on Euler parameterization and coprime factorization theory, a dynamic compensation architecture based on residual generator is designed. The transfer function of the dynamic compensation controller Qcom (S) is obtained by model matching to realize the steady-state operation of parallel inverters and the improvement of system power quality. Finally, the effectiveness of the proposed architecture is verified on the experimental platform. © 2023 IEEE.

关键词： Power quality

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：