arXiv - CS - Systems and Control最新文献_第9页

Distributed Cooperative Formation Control of Nonlinear Multi-Agent System (UGV) Using Neural Network 利用神经网络实现非线性多代理系统（UGV）的分布式合作编队控制

arXiv - CS - Systems and Control Pub Date : 2024-03-20 DOI: arxiv-2403.13473

Si Kheang Moeurn

{"title":"Distributed Cooperative Formation Control of Nonlinear Multi-Agent System (UGV) Using Neural Network","authors":"Si Kheang Moeurn","doi":"arxiv-2403.13473","DOIUrl":"https://doi.org/arxiv-2403.13473","url":null,"abstract":"The paper presented in this article deals with the issue of distributed\u0000cooperative formation of multi-agent systems (MASs). It proposes the use of\u0000appropriate neural network control methods to address formation requirements\u0000(uncertainties dynamic model). It considers an adaptive leader-follower\u0000distributed cooperative formation control based on neural networks (NNs)\u0000developed for a class of second-order nonlinear multi-agent systems and neural\u0000networks Neural networks are used to compute system data that inputs layer\u0000(position, velocity), hidden layers, and output layer. Through collaboration\u0000between leader-follower approaches and neural networks with complex systems or\u0000complex conditions receive an effective cooperative formation control method.\u0000The sufficient conditions for the system stability were derived using Lyapunov\u0000stability theory, graph theory, and state space methods. By simulation, the\u0000results of this study can be obtained from the main data of the multi-agent\u0000system in formation control and verified that the system can process\u0000consistency, stability, reliability, and accuracy in cooperative formation.","PeriodicalId":501062,"journal":{"name":"arXiv - CS - Systems and Control","volume":"4 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140196615","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Lattice piecewise affine approximation of explicit model predictive control with application to satellite attitude control 应用于卫星姿态控制的显式模型预测控制的晶格片状仿射近似法

arXiv - CS - Systems and Control Pub Date : 2024-03-20 DOI: arxiv-2403.13601

Zhengqi Xu, Jun Xu, Ai-Guo Wu, Shuning Wang

{"title":"Lattice piecewise affine approximation of explicit model predictive control with application to satellite attitude control","authors":"Zhengqi Xu, Jun Xu, Ai-Guo Wu, Shuning Wang","doi":"arxiv-2403.13601","DOIUrl":"https://doi.org/arxiv-2403.13601","url":null,"abstract":"Satellite attitude cotrol is a crucial part of aerospace technology, and\u0000model predictive control(MPC) is one of the most promising controllers in this\u0000area, which will be less effective if real-time online optimization can not be\u0000achieved. Explicit MPC converts the online calculation into a table lookup\u0000process, however the solution is difficult to obtain if the system dimension is\u0000high or the constraints are complex. The lattice piecewise affine(PWA) function\u0000was used to represent the control law of explicit MPC, although the online\u0000calculation complexity is reduced, the offline calculation is still prohibitive\u0000for complex problems. In this paper, we use the sample points in the feasible\u0000region with their corresponding affine functions to construct the lattice PWA\u0000approximation of the optimal MPC controller designed for satellite attitude\u0000control. The asymptotic stability of satellite attitude control system under\u0000lattice PWA approximation has been proven, and simulations are executed to\u0000verify that the proposed method can achieve almost the same performance as\u0000linear online MPC with much lower online computational complexity and use less\u0000fuel than LQR method.","PeriodicalId":501062,"journal":{"name":"arXiv - CS - Systems and Control","volume":"86 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140196617","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Bayesian Physics-informed Neural Networks for System Identification of Inverter-dominated Power Systems 贝叶斯物理信息神经网络用于逆变器主导型电力系统的系统识别

arXiv - CS - Systems and Control Pub Date : 2024-03-20 DOI: arxiv-2403.13602

Simon Stock, Davood Babazadeh, Christian Becker, Spyros Chatzivasileiadis

{"title":"Bayesian Physics-informed Neural Networks for System Identification of Inverter-dominated Power Systems","authors":"Simon Stock, Davood Babazadeh, Christian Becker, Spyros Chatzivasileiadis","doi":"arxiv-2403.13602","DOIUrl":"https://doi.org/arxiv-2403.13602","url":null,"abstract":"While the uncertainty in generation and demand increases, accurately\u0000estimating the dynamic characteristics of power systems becomes crucial for\u0000employing the appropriate control actions to maintain their stability. In our\u0000previous work, we have shown that Bayesian Physics-informed Neural Networks\u0000(BPINNs) outperform conventional system identification methods in identifying\u0000the power system dynamic behavior under measurement noise. This paper takes the\u0000next natural step and addresses the more significant challenge, exploring how\u0000BPINN perform in estimating power system dynamics under increasing uncertainty\u0000from many Inverter-based Resources (IBRs) connected to the grid. These\u0000introduce a different type of uncertainty, compared to noisy measurements. The\u0000BPINN combines the advantages of Physics-informed Neural Networks (PINNs), such\u0000as inverse problem applicability, with Bayesian approaches for uncertainty\u0000quantification. We explore the BPINN performance on a wide range of systems,\u0000starting from a single machine infinite bus (SMIB) system and 3-bus system to\u0000extract important insights, to the 14-bus CIGRE distribution grid, and the\u0000large IEEE 118-bus system. We also investigate approaches that can accelerate\u0000the BPINN training, such as pretraining and transfer learning. Throughout this\u0000paper, we show that in presence of uncertainty, the BPINN achieves orders of\u0000magnitude lower errors than the widely popular method for system identification\u0000SINDy and significantly lower errors than PINN, while transfer learning helps\u0000reduce training time by up to 80 %.","PeriodicalId":501062,"journal":{"name":"arXiv - CS - Systems and Control","volume":"42 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140196735","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Integrating Large Language Models for Severity Classification in Traffic Incident Management: A Machine Learning Approach 在交通事故管理中整合大型语言模型进行严重程度分类：机器学习方法

arXiv - CS - Systems and Control Pub Date : 2024-03-20 DOI: arxiv-2403.13547

Artur Grigorev, Khaled Saleh, Yuming Ou, Adriana-Simona Mihaita

{"title":"Integrating Large Language Models for Severity Classification in Traffic Incident Management: A Machine Learning Approach","authors":"Artur Grigorev, Khaled Saleh, Yuming Ou, Adriana-Simona Mihaita","doi":"arxiv-2403.13547","DOIUrl":"https://doi.org/arxiv-2403.13547","url":null,"abstract":"This study evaluates the impact of large language models on enhancing machine\u0000learning processes for managing traffic incidents. It examines the extent to\u0000which features generated by modern language models improve or match the\u0000accuracy of predictions when classifying the severity of incidents using\u0000accident reports. Multiple comparisons performed between combinations of\u0000language models and machine learning algorithms, including Gradient Boosted\u0000Decision Trees, Random Forests, and Extreme Gradient Boosting. Our research\u0000uses both conventional and language model-derived features from texts and\u0000incident reports, and their combinations to perform severity classification.\u0000Incorporating features from language models with those directly obtained from\u0000incident reports has shown to improve, or at least match, the performance of\u0000machine learning techniques in assigning severity levels to incidents,\u0000particularly when employing Random Forests and Extreme Gradient Boosting\u0000methods. This comparison was quantified using the F1-score over uniformly\u0000sampled data sets to obtain balanced severity classes. The primary contribution\u0000of this research is in the demonstration of how Large Language Models can be\u0000integrated into machine learning workflows for incident management, thereby\u0000simplifying feature extraction from unstructured text and enhancing or matching\u0000the precision of severity predictions using conventional machine learning\u0000pipeline. The engineering application of this research is illustrated through\u0000the effective use of these language processing models to refine the modelling\u0000process for incident severity classification. This work provides significant\u0000insights into the application of language processing capabilities in\u0000combination with traditional data for improving machine learning pipelines in\u0000the context of classifying incident severity.","PeriodicalId":501062,"journal":{"name":"arXiv - CS - Systems and Control","volume":"24 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140196506","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Control-Recoverable Added-Noise-based Privacy Scheme for LQ Control in Networked Control Systems 网络控制系统中 LQ 控制的可控可恢复附加噪声隐私方案

arXiv - CS - Systems and Control Pub Date : 2024-03-20 DOI: arxiv-2403.13346

Xuening Tang, Xianghui Cao, Wei Xing Zheng

{"title":"A Control-Recoverable Added-Noise-based Privacy Scheme for LQ Control in Networked Control Systems","authors":"Xuening Tang, Xianghui Cao, Wei Xing Zheng","doi":"arxiv-2403.13346","DOIUrl":"https://doi.org/arxiv-2403.13346","url":null,"abstract":"As networked control systems continue to evolve, ensuring the privacy of\u0000sensitive data becomes an increasingly pressing concern, especially in\u0000situations where the controller is physically separated from the plant. In this\u0000paper, we propose a secure control scheme for computing linear quadratic\u0000control in a networked control system utilizing two networked controllers, a\u0000privacy encoder and a control restorer. Specifically, the encoder generates two\u0000state signals blurred with random noise and sends them to the controllers,\u0000while the restorer reconstructs the correct control signal. The proposed design\u0000effectively preserves the privacy of the control system's state without\u0000sacrificing the control performance. We theoretically quantify the\u0000privacy-preserving performance in terms of the state estimation error of the\u0000controllers and the disclosure probability. Additionally, the proposed\u0000privacy-preserving scheme is also proven to satisfy differential privacy.\u0000Moreover, we extend the proposed privacy-preserving scheme and evaluation\u0000method to cases where collusion between two controllers occurs. Finally, we\u0000verify the validity of our proposed scheme through simulations.","PeriodicalId":501062,"journal":{"name":"arXiv - CS - Systems and Control","volume":"69 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140196620","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Safety-Aware Reinforcement Learning for Electric Vehicle Charging Station Management in Distribution Network 面向配电网络电动汽车充电站管理的安全意识强化学习

arXiv - CS - Systems and Control Pub Date : 2024-03-20 DOI: arxiv-2403.13236

Jiarong Fan, Ariel Liebman, Hao Wang

{"title":"Safety-Aware Reinforcement Learning for Electric Vehicle Charging Station Management in Distribution Network","authors":"Jiarong Fan, Ariel Liebman, Hao Wang","doi":"arxiv-2403.13236","DOIUrl":"https://doi.org/arxiv-2403.13236","url":null,"abstract":"The increasing integration of electric vehicles (EVs) into the grid can pose\u0000a significant risk to the distribution system operation in the absence of\u0000coordination. In response to the need for effective coordination of EVs within\u0000the distribution network, this paper presents a safety-aware reinforcement\u0000learning (RL) algorithm designed to manage EV charging stations while ensuring\u0000the satisfaction of system constraints. Unlike existing methods, our proposed\u0000algorithm does not rely on explicit penalties for constraint violations,\u0000eliminating the need for penalty coefficient tuning. Furthermore, managing EV\u0000charging stations is further complicated by multiple uncertainties, notably the\u0000variability in solar energy generation and energy prices. To address this\u0000challenge, we develop an off-policy RL algorithm to efficiently utilize data to\u0000learn patterns in such uncertain environments. Our algorithm also incorporates\u0000a maximum entropy framework to enhance the RL algorithm's exploratory process,\u0000preventing convergence to local optimal solutions. Simulation results\u0000demonstrate that our algorithm outperforms traditional RL algorithms in\u0000managing EV charging in the distribution network.","PeriodicalId":501062,"journal":{"name":"arXiv - CS - Systems and Control","volume":"51 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140196622","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Charged Momentum: Electric Vehicle Surge in India's 2023 Landscape 充电动力：2023 年印度电动汽车的发展态势

arXiv - CS - Systems and Control Pub Date : 2024-03-20 DOI: arxiv-2403.13373

Rahul Wagh

引用次数: 0

Augmented Labeled Random Finite Sets and Its Application to Group Target Tracking 增强标记随机有限集及其在群体目标追踪中的应用

arXiv - CS - Systems and Control Pub Date : 2024-03-20 DOI: arxiv-2403.13562

Chaoqun Yang, Mengdie Xu, Xiaowei Liang, Heng Zhang, Xianghui Cao

{"title":"Augmented Labeled Random Finite Sets and Its Application to Group Target Tracking","authors":"Chaoqun Yang, Mengdie Xu, Xiaowei Liang, Heng Zhang, Xianghui Cao","doi":"arxiv-2403.13562","DOIUrl":"https://doi.org/arxiv-2403.13562","url":null,"abstract":"This paper addresses the problem of group target tracking (GTT), wherein\u0000multiple closely spaced targets within a group pose a coordinated motion. To\u0000improve the tracking performance, the labeled random finite sets (LRFSs) theory\u0000is adopted, and this paper develops a new kind of LRFSs, i.e., augmented LRFSs,\u0000which introduces group information into the definition of LRFSs. Specifically,\u0000for each element in an LRFS, the kinetic states, track label, and the\u0000corresponding group information of its represented target are incorporated.\u0000Furthermore, by means of the labeled multi-Bernoulli (LMB) filter with the\u0000proposed augmented LRFSs, the group structure is iteratively propagated and\u0000updated during the tracking process, which achieves the simultaneously\u0000estimation of the kinetic states, track label, and the corresponding group\u0000information of multiple group targets, and further improves the GTT tracking\u0000performance. Finally, simulation experiments are provided, which well\u0000demonstrates the effectiveness of the labeled multi-Bernoulli filter with the\u0000proposed augmented LRFSs for GTT tracking.","PeriodicalId":501062,"journal":{"name":"arXiv - CS - Systems and Control","volume":"7 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140196508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

3D Directed Formation Control with Global Shape Convergence using Bispherical Coordinates 使用双球面坐标进行全局形状收敛的三维定向编队控制

arXiv - CS - Systems and Control Pub Date : 2024-03-20 DOI: arxiv-2403.13609

Omid Mirzaeedodangeh, Farhad Mehdifar, Dimos V. Dimarogonas

引用次数: 0

A Log-domain Interior Point Method for Convex Quadratic Games 凸二次方博弈的对数域内点法

arXiv - CS - Systems and Control Pub Date : 2024-03-20 DOI: arxiv-2403.13290

Bingqi Liu, Dominic Liao-McPherson

引用次数: 0