检索结果-内蒙古大学图书馆

Reinforcement learning for optimal tracking of large-scale systems with multitime scales

Science China(Information Sciences) 2023年第7期66卷 5-29页

作者： Jinna LI Hao NIE Tianyou CHAI Frank L.LEWIS School of Information and Control Engineering Liaoning Petrochemical University State Key Laboratory of Synthetical Automation for Process Industries Northeastern University UTA Research Institute University of Texas at Arlington

This paper aims to solve an optimal tracking control(OTC) problem of large-scale systems with multitime scales and coupled subsystems using singular perturbation(SP) theory and reinforcement learning(RL) techniques. A considerable contribution of this paper is the development of a data-driven SP-based RL method for the OTC of unknown large-scale systems with multitime scales. To achieve this, a multitime scale tracking problem was decomposed into a linear quadratic tracker problem for slow subsystems and a dynamical game problem for fast subsystems using the SP theory. Then, the distributed composite feedback controllers were found using a distributed off-policy integral RL algorithm that uses only measured data from the system in real time. Thus, the operational index can follow its prescribed target value via an approximately optimal approach. Theoretical analysis and proof are presented to demonstrate that the sum of the performances of reduced-order subsystems is approximately equal to the performance of the original large-scale system. Finally, numerical and practical examples are provided to validate the effectiveness of the proposed method.

关键词： reinforcement learning singular perturbation multitime scales data-driven control

来源：评论

学校读者我要写书评

暂无评论

Distributed formation control of multiple aerial vehicles based on guidance route

引用

Chinese Journal of Aeronautics 2023年第3期36卷 368-381页

作者： Jinyong CHEN Rui ZHOU Guibin SUN Qingwei LI Ning ZHANG School of Automation Science and Electrical Engineering Beihang UniversityBeijing 100191China Science and Technology on Information Systems Engineering Laboratory Nanjing 210007China AVIC Xi’an Flight Automatic Control Research Institute Xi’an 710076China

Formation control of fixed-wing aerial vehicles is an important yet rarely addressed problem because of their complex dynamics and various motion constraints,such as nonholonomic and velocity *** guidance-route-based strategy has been demonstrated to be applicable to fixed-wing ***,it requires a global coordinator and there exists control lag,due to its own *** this reason,this paper presents a fully distributed guidance-route-based formation approach to address the aforementioned ***,a hop-count scheme is introduced to achieve distributed implementation,in which each aircraft chooses a neighbor with the minimum hop-count as a reference to generate its guidance route using only local ***,the model predictive control algorithm is employed to eliminate the control lag and achieve precise formation shape *** addition,the stall protection and collision avoidance are also ***,three numerical simulations demonstrate that our proposed approach can implement precise formation shape control of fixed-wing aircraft in a fully distributed manner.

关键词： Formation shape control Fixed-wing aircraft Guidance route Hop-count estimation Model predictive control

来源：评论

学校读者我要写书评

暂无评论

Graph-guided masked autoencoder for process anomaly detection

引用

process Safety and Environmental Protection 2024年 186卷 1345-1357页

作者： Jia, Mingwei Xu, Danya Yang, Tao Yao, Yuan Liu, Yi Institute of Process Equipment and Control Engineering Zhejiang University of Technology Hangzhou310023 China State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Shenyang110819 China Department of Chemical Engineering National Tsing Hua University Hsinchu300044 Taiwan

Developing an accurate and reliable anomaly detection model is of great significance for safe operation in the process industry. To minimize false positives, it is crucial to accurately model the intricate topological and nonlinear connections among variables. In this study, a process anomaly detection method based on graph-guided masked autoencoder (GGMAE) is proposed by introducing the concept of the graph to the process industry. GGMAE first constructs a topology graph according to the process flowchart. Then, the patch and mask mechanism forces GGMAE to learn the process intrinsic information to reconstruct the input according to the variable topological relationship and temporal characteristics. Additionally, Kullback-Leibler divergence is used as the loss to ensure that the distribution of input and output is consistent. Experiments on two publicly available anomaly detection benchmarks demonstrate the superiority of GGMAE over existing methods. Visualization of the results demonstrates the physically compliant reconstruction logic. © 2024 The Institution of Chemical Engineers

关键词： Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

Bayesian Filtering for High-Dimensional State-Space Models With State Partition and Error Compensation

引用

IEEE/CAA Journal of automatica Sinica 2024年第5期11卷 1239-1249页

作者： Ke Li Shunyi Zhao Biao Huang Fei Liu Key Laboratory of Advanced Process Control for Light Industry(Ministry of Education) Institute of AutomationJiangnan UniversityWuxi 214122China Department of Chemical and Materials Engineering University of AlbertaEdmonton AB T6G 2G6Canada

In the era of exponential growth of data availability,the architecture of systems has a trend toward high dimensionality,and directly exploiting holistic information for state inference is not always computationally *** paper proposes a novel Bayesian filtering algorithm that considers algorithmic computational cost and estimation accuracy for high-dimensional linear *** high-dimensional state vector is divided into several blocks to save computation resources by avoiding the calculation of error covariance with immense *** that,two sequential states are estimated simultaneously by introducing an auxiliary variable in the new probability space,mitigating the performance degradation caused by state ***,the computational cost and error covariance of the proposed algorithm are analyzed analytically to show its distinct features compared with several existing *** results illustrate that the proposed Bayesian filtering can maintain a higher estimation accuracy with reasonable computational cost when applied to high-dimensional linear systems.

关键词： filtering estimation error

来源：评论

学校读者我要写书评

暂无评论

An Optimization Method for the Swing of a Quadrotor Flight Suspension System 25

An Optimization Method for the Swing of a Quadrotor Flight S...

引用

25th IEEE China Conference on System Simulation Technology and its Application, CCSSTA 2024

作者： Yang, Sen Zang, Xinlian Tong, Jigang Xie, Tian Jin, Xin Tianjin University of Technology The School of Electrical Engineering and Automation Tianjin China AVIC Xi'an Flight Automatic Control Research Institute National key laboratory of science and technology on aircraft control Xi'an China

ISBN: (纸本)9798350366600

The aim of this study is to explore optimization control methods for suspension systems during the flight of quadcopters. The traditional control method of suspension system has some problems such as oscillation and instability, so an optimization scheme based on Gaussian pseudo-general algorithm is proposed in this paper. Through numerical simulation and actual flight test, the effectiveness of the optimization algorithm in improving system stability and reducing oscillation amplitude is verified. At the same time, this study conducted a comparative analysis of the suspension system of the quadrotor vehicle and the free fall body, and the optimization algorithm is found to be feasible and can well improve the control accuracy and stability of the system, and provide strong support for the safety of the quadrotor vehicle in various application scenarios. © 2024 IEEE.

关键词： Flight testing

来源：评论

学校读者我要写书评

暂无评论

A Enhanced Feature Extraction for 3D Object Detection 25

A Enhanced Feature Extraction for 3D Object Detection

引用

25th IEEE China Conference on System Simulation Technology and its Application, CCSSTA 2024

作者： Meng, Qingao Tong, Jigang Yang, Sen Xie, Tian Jin, Xin Tianjin University of Technology The School of Electrical Engineering and Automation Tianjin China Avic Xi'an Flight Automatic Control Research Institute National Key Laboratory of Science and Technology on Aircraft Control Xi'an China

ISBN: (纸本)9798350366600

3D object detection is crucial for autopilot and augmented reality applications. However, accurate feature extraction is the key to achieve high-precision detection, but traditional methods face the challenges of information loss and computational inefficiency. To address this, we propose an enhanced feature extraction method. We first divide the 3D space into voxel grids of varying scales. Then, using 3D sparse convolution, we extract features which are compressed along the Z axis to form BEV (Bird's Eye View) feature images. An efficient multi-scale attention mechanism is then employed to enhance these BEV features. The refined feature map is subsequently located and classified through a CT-stacking module and RPN (Region Proposal Network). Voxel features from the proposal and 3D backbone are passed through a voxel RoI pool to directly extract RoI features from the 3D features for further refinement. Finally, a cascaded attention network refines the proposal and strengthens confidence prediction. Experiments on test set validate our method effectiveness, showing improved accuracy over traditional voxel methods. Overall, this enhanced feature extraction technique offers a promising solution for 3D target detection, particularly in scenarios demanding high real-time performance and precision. © 2024 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Intelligent Flight control for Cruise Vehicle Based on Backstepping Design and Reinforcement Learning

Intelligent Flight Control for Cruise Vehicle Based on Backs...

引用

2023 China automation Congress, CAC 2023

作者： Liu, Xiaodong Jia, Chenhui Yang, Qingjun Wu, Tiancai Zhang, Yu Beijing Aerospace Automatic Control Institute National Key Laboratory of Science and Technology on Aerospace Intelligence Control Beijing China Beijing Aerospace Automatic Control Institute Beijing China School of Automation Science and Electrical Engineering Beihang University Beijing China

ISBN: (纸本)9798350303759

Aiming at velocity/height tracking control and attitude stabilization for the longitudinal flight of cruise vehicle, the research of intelligent flight control method is carried out by combining the backstepping control theory and the deep reinforcement learning method. Firstly, the longitudinal flight control system of the vehicle is decomposed into velocity and height subsystems, and the control law is designed based on backstepping control theory. Then, aiming at the problems that the selection of the multiple control gains will directly affect the control performance of the closed-loop system and the adjustment process of control gains is cumbersome, the twin delayed deep deterministic policy gradient algorithm (TD3) will be adopted to train the cruise vehicle, so that it can determine the control gains online according to the current flight states. The numerical simulation results show that, the proposed intelligent flight control method can guarantee the high-precision tracking with respect to height and velocity commands, and also realize the attitude stability of cruise vehicle. © 2023 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Data-Driven control of Distributed Event-Triggered Network Systems

引用

IEEE/CAA Journal of automatica Sinica 2023年第2期10卷 351-364页

作者： Xin Wang Jian Sun Gang Wang Frank Allgower Jie Chen the National Key Laboratory of Autonomous Intelligent Unmanned Systems School of AutomationBeijing Institute of TechnologyBeijing 100081 the Beijing Institute of Technology Chongqing Innovation Center Chongqing 401120China IEEE the Institute for Systems Theory and Automatic Control University of StuttgartStuttgart 70550Germany the Department of Control Science and Engineering Tongji UniversityShanghai 201804 the National Key Laboratory of Autonomous Intelligent Unmanned Systems Beijing Institute of TechnologyBeijing 100081China

The present paper deals with data-driven event-triggered control of a class of unknown discrete-time interconnected systems(*** systems).To this end,we start by putting forth a novel distributed event-triggering transmission strategy based on periodic sampling,under which a model-based stability criterion for the closed-loop network system is derived,by leveraging a discrete-time looped-functional *** the model-based criterion with a data-driven system representation recently developed in the literature,a purely data-driven stability criterion expressed in the form of linear matrix inequalities(LMIs)is ***,the data-driven stability criterion suggests a means for co-designing the event-triggering coefficient matrix and the feedback control gain matrix using only some offline collected state-input ***,numerical results corroborate the efficacy of the proposed distributed data-driven event-triggered network system(ETS)in cutting off data transmissions and the co-design procedure.

关键词： Data-driven control distributed event-triggered network system(ETS) linear matrix inequalitie(LMI) looped-functional stability

来源：评论

学校读者我要写书评

暂无评论

Graph convolutional network soft sensor for process quality prediction

引用

Journal of process control 2023年第1期123卷 12-25页

作者： Jia, Mingwei Xu, Danya Yang, Tao Liu, Yi Yao, Yuan Institute of Process Equipment and Control Engineering Zhejiang University of Technology Hangzhou310023 China State Key Laboratory of Synthetical Automation for Process Industries Northeastern University Shenyang110819 China Department of Chemical Engineering National Tsing Hua University Hsinchu30013 Taiwan

The nonlinear time-varying characteristics of the process industry can be modeled using numerous data-driven soft sensor methods. However, the intrinsic relationships among the variables, especially the localized spatial–temporal correlations that shed light on model behavior, have received little attention. In this study, a soft sensor based on a graph convolutional network is constructed by introducing the concept of graph to process modeling. The focus is on obtaining localized spatial–temporal correlations that aid in comprehending the intricate interactions among the variables included in the soft sensor. The model is trained by considering the regularization terms and it learns distinctive localized spatial–temporal correlations in an end-to-end manner. Furthermore, long-term dependence is established via temporal convolution. Thus, both the localized spatial–temporal correlations and time-series properties are captured. The feasibility of the proposed soft sensor is illustrated using two fermentation processes. The localized spatial–temporal correlations of this case study are visualized, and they demonstrate that the soft sensor is not a black-box model;instead, it is consistent with process knowledge. © 2023 Elsevier Ltd

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Latent-Enhanced Variational Adversarial Active Learning Assisted Soft Sensor

引用

IEEE Sensors Journal 2023年第14期23卷 15762-15772页

作者： Dai, Yun Yang, Chao Liu, Yi Yao, Yuan Zhejiang University of Technology Institute of Process Equipment and Control Engineering Hangzhou310023 China Northeastern University State Key Laboratory of Synthetical Automation for Process Industries Shenyang110819 China National Tsing Hua University Department of Chemical Engineering Hsinchu300044 Taiwan

As the acquisition of variables that measure quality is typically challenging, labeled samples for building a model for soft sensors are often inadequate. Additionally, owing to the installation of redundant sensors, high-dimensional process data with strong correlations are acquired. Therefore, in this study, a sample selection strategy for active learning (AL), referred to as latent-enhanced variational adversarial AL (LVAAL), is developed to enhance quality prediction performance with limited labeled data. The LVAAL method uses a minimax game to explore the latent representation of the original data. Specifically, a latent-enhanced variational autoencoder (LEVAE) is used to deceive the adversarial network into predicting all samples, both labeled and unlabeled, as labeled. However, the adversarial network attempts to indicate the difference between latent representations. Subsequently, Gaussian process regression (GPR) is adopted as a base model for LVAAL. With the proposed LVAAL selection strategy, informative unlabeled samples are incrementally selected, thereby enhancing the prediction performance. The application results for a numerical case and an industrial example demonstrate the advantages of LVAAL compared with existing AL strategies. © 2001-2012 IEEE.

关键词： Gaussian distribution

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：