检索结果-内蒙古大学图书馆

Modeling, Positioning, and Deep Reinforcement Learning Path Following control of Scaled Robotic Vehicles: Design and Experimental Validation

引用

IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 2025年 22卷 9856-9871页

作者： Caponio, Carmine Stano, Pietro Carli, Raffaele Olivieri, Ignazio Ragone, Daniele Sorniotti, Aldo Gruber, Patrick Montanaro, Umberto Univ Surrey Sch Mech Engn Sci Guildford GU2 7XH England Polytech Bari Dept Elect & Informat Engn I-70125 Bari Italy Polytech Univ Turin Dept Mech & Aerosp Engn I-10129 Turin Italy

Mobile robotic systems serve as versatile platforms for diverse indoor applications, ranging from warehousing and manufacturing to test benches dedicated to evaluating automated driving (AD) functions. In AD systems, the path following (PF) layer is responsible for defining steering commands to follow the reference path. Recently explored approaches involve artificial intelligence-based methods, such as Deep Reinforcement Learning (DRL). Despite their promising performance, these controllers still suffer from time-consuming training phases and may experience performance degradation when deviating from training conditions. To address these challenges, this paper proposes novel DRL controllers addressing the simulation-to-reality gap in unknown scenarios by: (i) training via an expert demonstrator which also speed up the learning phase;and (ii) a weight adaptation strategy for the resulting neural network (NN) to strengthen controller robustness and enhance PF performance. In addition, an experimentally validated vehicle model is used for training the proposed DRL algorithm and as a model for a federated extended Kalman filter (FEKF) system employed for sensor fusion in vehicle localisation. The proposed DRL-based PF controllers are experimentally evaluated through key performance indicators across multiple maneuvers not considered during training, and it is shown that they outperform benchmarking model-based controllers from the literature. Note to Practitioners-This paper presents a comprehensive toolchain for controlling mobile robots, which includes: (i) a simple yet effective two-stage least-square approach for parameter identification of the longitudinal and lateral dynamics of scaled robotic vehicles;(ii) the utilisation of a no-reset FEKF to enhance positioning leveraging all sensors commonly available on scaled robotic vehicles;(iii) the inclusion of an expert demonstrator to expedite the training phase and address the simulation-to-reality gap resulting from

关键词： Training Robots Robot sensing systems Vehicle dynamics Accuracy Artificial neural networks Laser radar Context modeling Robustness Kalman filters Automated vehicles vehicle modelling position estimation path following deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Single Model Scheme-based Smith Predictor for the Mitigation of the effects of network Imperfections in Consensus control of Multi Agent systems 10

Single Model Scheme-based Smith Predictor for the Mitigation...

引用

10th International conference on control, Decision and Information Technologies (CoDIT)

作者： Solyman, Loaie Elbadawyl, Ayman Meroth, Ansgar German Univ Cairo Fac Engn & Mat Sci Mechatron Engn Dept Cairo Egypt Heilbronn Univ Automot Syst Engn Dept Heilbronn Germany

ISBN: (纸本)9798350373981;9798350373974

This paper addresses the problem of network modelling and its application in consensus control of networked Multi Agent systems (NMASs). These NMASs experience network imperfections such as variable time delays and packet dropouts, where some researchers sought a model that predicts the same behaviour of these imperfections. Earlier studies had two models predicting the behaviour of each network imperfection independently in a scheme called the Two Independent Model Scheme (TIMS). A new approach of modelling the behaviour of the network, based on the Semi-Continuous Hidden Markov Model (SCHMM), is introduced in this paper. This was achieved through introducing a Dirac-delta function alongside the Gaussian distributions in the SCHMM. The new model predicts time delays and packet dropouts, in a homogeneous manner, in a Single Model Scheme (SMS). A Smith Predictor (SP) scheme was implemented for the consensus control of NMASs to mitigate the effects of network limitations. The numerical analysis shows that the new model is more accurate in representing the network limitations, and shows that its performance is better at mitigating the effects of network imperfections in the consensus control problem.

关键词： Delta functions

来源：评论

学校读者我要写书评

暂无评论

Model Predictive control for Offset-free Tracking: a new data-driven application for Secondary Voltage Regulation on Transmission network

Model Predictive Control for Offset-free Tracking: a new dat...

引用

International conference on Smart Energy systems and Technologies (SEST) - Driving the Advances for Future Electrification

作者： Acerbi, Federica Piccagli, Davide Stefano Pozzi, Fabio Sciarpa, Luca Giannuzzi, Giorgio Maria Pisani, Cosimo Bruno, Gianluca CESI SpA Milan Italy Terna SpA Rome Italy

ISBN: (纸本)9798350386509;9798350386493

This paper investigates the evolution of Secondary Voltage Regulation (SVR) in response to the increasing penetration of enewable energy sources (RES) within power systems. Traditional SVR, historically reliant on fossil fuel-based plants, faces challenges stemming from their under-utilization and intermittent operation. As conventional power plants decline, emerging resources such as synchronous compensators and STATCOMS offer voltage control capabilities without compromising RES integration, prompting the need for a redesigned control system to effectively harness their capabilities and optimize voltage regulation performance in an increasingly dynamic network environment. In this evolving scenario, a review of SVR literature reveals a shift towards data-driven methodologies, leveraging real-world data for improved control strategies. To address these challenges, a new Secondary Voltage Regulator is proposed based on a data-driven Model Predictive control (MPC) approach, designed for offset-free tracking. The suggested approach, known for its proficiency in tracking, has been adjusted to provide an implementation suitable for the Italian transmission system. Field tests conducted on Sicilian transmission network validate the effectiveness of the MPC-based controller under real-world conditions, filling an important gap in understanding its performance and applicability in transmission systems.

关键词： Data-driven Modeling Model Predictive control Reactive Power Management Secondary Voltage control

来源：评论

学校读者我要写书评

暂无评论

Security Tracking control for Discrete-Time Stochastic Delayed networked systems With Cyber-Attacks by Dynamic Event-Triggered Approach 36

Security Tracking Control for Discrete-Time Stochastic Delay...

引用

36th Chinese control and Decision conference (CCDC)

作者： Xu, Shiqiang Yang, Jun Wu, Qiong Southwest Minzu Univ Coll Elect Engn Key Lab Elect & Informat Engn State Ethn Affairs Chengdu Peoples R China

ISBN: (纸本)9798350387780;9798350387797

The purpose of this paper is to develop a security tracking controller for discrete-time delayed stochastic network control systems (SNCSs) that are vulnerable to stochastic cyber attacks. The presented strategy, called the Dynamic Event-Triggered Communication Approach (DETCA), adapts the volume of data transmitted across the network based on fluctuations in tracking error system, while concurrently maintaining the desired level of the tracking performance. These stochastic cyber attacks encompass deception attacks and denial-of-service (DoS), which may occur during signal transmission over the network. Based on Lyapunov stability theory, a sufficient condition is provided for guaranteeing the asymptotic stability of the tracking error system. The linear matrix inequalities (LMIs) are employed to solve the event-triggering parameters and the tracking controller gain. The efficacy of the proposed theoretical results is lastly demonstrated by a simulation example provided.

关键词： Discrete-time stochastic delayed systems networked controll systems Dynamic event-triggered communication approach (DETCA) Cyber attacks

来源：评论

学校读者我要写书评

暂无评论

SDN-based L4S Congestion control in Beyond 5G 25

SDN-based L4S Congestion Control in Beyond 5G

引用

IEEE 25th International conference on High performance Switching and Routing (HPSR)

作者： Messaoudi, Sofiane Ksentini, Adlen Messaoudi, Franck Bonnet, Christian Eurecom Inst Sophia Antipolis France OpenAirInterface Alliance Sophia Antipolis France

ISBN: (纸本)9798350363869;9798350363852

This paper describes the SDN-based L4S solution, a congestion control algorithm designed to improve QoS in 5G and Beyond networks. Inspired by the IETF specifications, our framework tackles challenges prevalent in immersive applications like video streaming and cloud gaming, such as ultra-low latency and packet loss. The proposed solution seamlessly integrates L4S techniques into SDN, thereby optimizing queue management within the transport network. Additionally, it employs Explicit Congestion Notification (ECN) to mark packets during congestion scenarios. This synergistic approach facilitates dynamic adjustments in transmission rates, enhancing the overall efficiency of the transport network, particularly in accommodating various classes of traffic. Evaluation results demonstrate that our solution outperforms the benchmarks by a substantial margin in terms of End-to-End latency, and packet loss.

关键词： Packet loss

来源：评论

学校读者我要写书评

暂无评论

Discrete Gain Scheduling Output Feedback control of Input-Saturated Linear Multi-agent systems 20th

Discrete Gain Scheduling Output Feedback Control of Input-Sa...

引用

20th Chinese Intelligent systems conference

作者： Gao, Xiangyu Zhong, Zefeng Xie, Weiwei Wang, Jianqiao Guangxi Normal Univ Sch Math & Stat Guilin 541004 GuangXi Peoples R China Tiangong Univ Sch Control Sci & Engn Tianjin 300387 Peoples R China

ISBN: (纸本)9789819786565;9789819786541;9789819786534

This paper considers the state observer-based discrete gain scheduling output feedback control problem for linear multi-agent systems with input saturation. The distributed discrete gain scheduling output feedback control method is proposed to studey bipartite consensus of input saturated linear multi-agent system with the directed network topology. It can be proved that the linear multi-agent system can achieve the bipartite consensus under the observer-based discrete gain scheduling feedback control protocol, which is obtained through solving the parameter Lyapunov equation. With the increases of discrete switching parameters, the observer-based discrete gain scheduling feedback control protocol can effectively improve the dynamic performance of multi-agent systems with input saturation. Finally, a numerical simulation is provided to show the validity of the theoretical results.

关键词： Linear multi-agent systems Input saturation Bipartite consensus Discrete gain scheduling output feedback control Analytical solution of the parametric Lyapunov equation

来源：评论

学校读者我要写书评

暂无评论

Dynamic, Reconfigurable and Green network Slice Admission control and Resource Allocation in the O-RAN Using Model Predictive control 23

Dynamic, Reconfigurable and Green Network Slice Admission Co...

引用

23rd International-Federation-for-Information-Processing (IFIP) networking conference (IFIP networking)

作者： Fryganiotis, Nikolaos Stai, Eleni Dimolitsas, Ioannis Zafeiropoulos, Anastasios Papavassiliou, Symeon Natl Tech Univ Athens Sch Elect & Comp Engn Inst Commun & Comp Syst ICCS Athens Greece

ISBN: (纸本)9798350390605;9783903176638

In the context of 5G, virtualization has transformed Radio Access network (RAN) architectures, enabling efficient resource utilization, flexibility and scalability of RAN deployments and operations. Within this paradigm, network slicing has emerged as a pivotal technique, enabling the creation of tailored virtual network instances to meet diverse service requirements. This study performs joint slice request admission control and optimal Virtual network Functions (VNFs) placement in O-RAN-enabled networks, subject to infrastructure and Quality-of-Service constraints. Contrary to existing schemes, emphasis is placed on two not yet deeply studied directions. The first is about handling future uncertainties on slice requests arrivals for which an iterative Model Predictive control (MPC)-based approach is proposed that leverages updated traffic forecasts for dynamic adaptation. The second relates to VNFs migration in the O-RAN modules to increase the slice acceptance ratio in an energy efficient way. Through performance evaluations and comparisons, we demonstrate the efficacy of the proposed MPC-based solution compared to other approaches.

关键词： network Slicing Radio Access network (RAN) Open RAN (O-RAN) network Function Virtualization Admission control Model Predictive control (MPC)

来源：评论

学校读者我要写书评

暂无评论

Research on the Physical Layer Vulnerability of Industrial control systems Based on Deterministic and Stochastic Petri Nets 24

Research on the Physical Layer Vulnerability of Industrial C...

引用

International conference on Algorithms, Software Engineering, and network Security (ASENS)

作者： Tang, Jin Shuang, Wei China Three Gorges Grp Corp Digital Management Ctr Wuhan 430000 Hubei Peoples R China China Three Gorges Corp Yangtze River Basin Hub Operat & Management Ctr Wuhan 430000 Hubei Peoples R China

ISBN: (纸本)9798400709784

With the development of modern industrialization, the network security issues of industrial control systems have become increasingly severe, necessitating enhanced research on the vulnerability mechanisms of industrial control network security. In response to the physical layer vulnerability issues in industrial control networks, a vulnerability analysis and performance quantification method based on deterministic and stochastic Petri net (DSPN) model is proposed. This method establishes a DSPN model using the Ethernet transmission mechanism as an example. Firstly, it qualitatively analyzes the possible abnormal states of the physical layer after being attacked. Then, it quantitatively analyzes the throughput and delay of the physical layer under abnormal states using DSPN tools and queuing theory. A simulation analysis was conducted on the throughput and transmission delay after the attack, and the result indicates that the Ethernet link negotiation mechanism is vulnerable. Timely wireless pulse injection can cause the port status to be abnormal. Although it does not lead to communication interruption, it will seriously affect communication performance.

关键词： Industrial control network Vulnerability analysis Deterministic and stochastic Petri nets Markov chain

来源：评论

学校读者我要写书评

暂无评论

Shared-control teleoperation methods for a cable-suspended dual-arm unmanned aerial manipulator

Shared-control teleoperation methods for a cable-suspended d...

引用

International conference on Unmanned Aircraft systems (ICUAS)

作者： Selvaggio, Mario Esposito, Federico Lippiello, Vincenzo Ruggiero, Fabio Univ Naples Federico II Dept Elect Engn & Informat Technol I-80125 Naples Italy

ISBN: (纸本)9798350357899;9798350357882

This paper introduces two shared-control teleoperation methods for remotely executing long-reach tasks with a cable-suspended dual-arm unmanned aerial manipulator. The proposed techniques aim to improve task performance and user experience during remote tasks involving interaction with the environment. Two application scenarios are envisioned: pushing against a flat surface to emulate in-contact inspection tasks of infrastructures, and object grasping to simulate debris removal in cluttered environments. The effectiveness of the two shared-control teleoperation methods is evaluated through a human-subjects study involving 10 participants commanding the simulated robot via a joystick interface. Statistical analysis demonstrates significant enhancements in task performance and system usability when using the proposed methods compared to standard teleoperation.

关键词： Remote control

来源：评论

学校读者我要写书评

暂无评论

Decentralized Cooperative Multi-Agent Deep Reinforcement Learning for Real-Time Optimization of Emulated Ad-Hoc Radio networks

Decentralized Cooperative Multi-Agent Deep Reinforcement Lea...

引用

2024 Military Communications conference

作者： Moehlenhof, Thies Jansen, Norman Fraunhofer FKIE Informat Technol Command & Control ITF Wachtberg Germany

ISBN: (纸本)9798350374247;9798350374230

This research explores the application of policy gradient methods in multi-agent reinforcement learning, augmented with offline reinforcement Learning techniques. The goal is to leverage these methods to improve communication of military command and control information systems (C2IS) in real-time emulated radio networks by the use of agents deployed on each network node in a decentralized manner. These agents have the task of improving the performance of the network by learning to adapt the local C2IS and communication services on their designated nodes to the prevailing dynamic network conditions. The proposed method is assessed in an emulated environment, where agents are trained to augment the network performance by controlling the transmission rate of a set of Blue-Force Tracking Services within an emulated ad-hoc radio network in real-time.

关键词： Multi-Agent Reinforcement Learning network Optimization Ad-Hoc Radio network Emulation Real-Time Reinforcement Learning

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：