检索结果-内蒙古大学图书馆

Inverse Reinforcement learning for Discrete-Time systems With data Dropouts

ieee TRANSACTIONS ON CYBERNETICS 2025年第4期55卷 1744-1757页

作者： Fan, Jialu Shi, Pengfei Xue, Wenqian Lian, Bosen Cui, Yunfang Lewis, Frank L. Northeastern Univ State Key Lab Synthet Automat Proc Ind & Int Joint Shenyang 110819 Peoples R China Univ Florida Elect & Comp Engn Dept Gainesville FL 32611 USA Auburn Univ Elect & Comp Engn Dept Auburn AL 36849 USA Univ Texas Arlington UTA Res Inst Ft Worth TX 76118 USA

This article proposes inverse reinforcement learning (IRL) algorithms for tracking control of linear networked control systems under random state dropouts during wireless transmission. The controlled system aims to track the optimal trajectory of a target system, despite the cost function governing the target's behaviors being unknown. The problem is complicated by random state dropouts occurring in two crucial scenarios: 1) the reception of the target's state and 2) feedback of the controlled system's states. Our approach enables the controlled system to infer the target's cost function and optimal control policy, thereby facilitating effective tracking. Specifically, we develop a model-based IRL algorithm that integrates the Smith predictor for state estimation. Then, we advance a state-dropout-aware inverse Q-learning algorithm that uses solely accessible system data, eliminating the need for system models. The theoretical validity of the proposed algorithms is rigorously established, and their practical effectiveness is validated through numerical simulations.

关键词： Cost function control systems Optimal control Target tracking Trajectory Prediction algorithms Wireless networks Stability analysis Fans Electronic mail data dropouts data-driven tracking control inverse reinforcement learning (IRL) reinforcement learning (RL)

来源：评论

学校读者我要写书评

暂无评论

Model-Based and data-driven Stochastic Hybrid control for Rumor Propagation in Dual-Layer Network

引用

ieee TRANSACTIONS ON COMPUTATIONAL SOCIAL systems 2025年第2期12卷 777-791页

作者： Zhong, Xiaojing Luo, Chaolong Deng, Feiqi Liu, Guiyun Li, Chunlei Hu, Zhipei Guangzhou Univ Sch Mech & Elect Engn Guangzhou 510006 Peoples R China South China Univ Technol Sch Automat Guangzhou 510641 Peoples R China Guangzhou Univ Inst New Media & Social Dev Huangpu Grad Sch Guangzhou 510006 Peoples R China

Toward exploring the positive impact of media debunking and random blocking on the spread of rumors, we discuss a stochastic hybrid control strategy that combines an individual and media debunking method, a continuous stochastic blocking method, and an impulse interruption method. Using stochastic analysis, the almost sure exponential stability of the controlled system is analyzed, along with the expression of control intensities. To balance rumor suppression, minimize control costs, and enhance the generality of control, a data-driven machine learning (ML) approach is developed to provide suboptimal control solutions. Numerical simulations based on two real-case datasets are carried out to validate the theoretical results and evaluate the potential impact of the model-based, data-driven stochastic hybrid control strategy.

关键词： data-driven ML control algorithm dual-layer network model rumor debunking method stochastic blocking method stochastic hybrid control data-driven ML control algorithm dual-layer network model rumor debunking method stochastic blocking method stochastic hybrid control

来源：评论

学校读者我要写书评

暂无评论

Event-Based Optimal Containment control for Constrained Multiagent systems Using Integral Reinforcement learning

引用

ieee TRANSACTIONS ON control OF NETWORK systems 2025年第1期12卷 609-619页

作者： Guo, Zijie Ren, Hongru Li, Hongyi Huang, Tingwen Guangdong Polytech Normal Univ Sch Elect & Informat Guangzhou 510665 Peoples R China Guangdong Univ Technol Sch Automat Guangdong Hong Kong Joint Lab Intelligent Decis & Guangzhou 510006 Guangdong Peoples R China Guangdong Univ Technol Guangdong Prov Key Lab Intelligent Decis & Coopera Guangzhou 510006 Peoples R China Southwest Univ Coll Elect & Informat Engn Chongqing 400715 Peoples R China Shenzhen Univ Adv Technol Fac Comp Sci & Control Engn Shenzhen 518055 Peoples R China

An optimal event-driven containment control problem is studied for partially unknown nonlinear multiagent systems with input constraints and state constraints. Its novelty lies in the optimization of the performance index while ensuring constraints handling abilities on states and inputs. First, an improved discounted cost function is constructed, and the state and input constraint information are encoded into the cost function by barrier functions and nonquadratic utility functions, respectively. Then, the approximate distributed optimal containment control policy is derived by an integral reinforcement learning (IRL)-based adaptive critic design, where the IRL technique can overcome the limitation of known drift dynamics in previous results. In critic neural networks learning, the weight tuning law is presented by virtue of the concurrent learning technique, which relaxes the persistence of excitation conditions by storing appropriate historical data. In order to reduce the amount of information transmitted through the controller-to-actuator channel, a containment error-dependent dynamic event-triggered mechanism is defined. Theoretical results indicate that signals in closed-loop systems driven by event-triggered optimal controllers are uniformly ultimately bounded, and Zeno behavior is avoided. Finally, the effectiveness of the developed method is illustrated by a simulation example on multiple single-link robot manipulators.

关键词： Event detection control systems Manipulators Reinforcement learning Artificial neural networks Resource management Cost function Aerospace electronics Vectors Safety Containment control event-triggered control input constraint integral reinforcement learning (IRL) state constraint

来源：评论

学校读者我要写书评

暂无评论

A data-driven Obstacle Avoidance Scheme for Redundant Robots With Unknown Structures

引用

ieee TRANSACTIONS ON INDUSTRIAL INFORMATICS 2025年第2期21卷 1793-1802页

作者： Xie, Zhengtai Liu, Mei Su, Zhenming Sun, Zhongbo Jin, Long Lanzhou Univ Sch Informat Sci & Engn Lanzhou 730000 Peoples R China Changchun Univ Technol Dept Control Engn Changchun 130012 Peoples R China

Redundant robots may undergo structural changes due to factors such as modifications, which pose challenges to their precise control and obstacle avoidance. To resolve this issue, this article proposes a data-driven obstacle avoidance (DDOA) scheme for redundant robots with unknown structures, which integrates obstacle avoidance control and structure learning. To ensure collision-free operations, an obstacle avoidance method for redundant robots is devised to maintain a safe distance from obstacles. Simultaneously, a data-driven learning equation is developed to estimate two Jacobian matrices of robots for obstacle avoidance and motion planning. A recurrent neural network (RNN) is then established to find the optimal solution to the DDOA scheme with theoretical analyses. Furthermore, we demonstrate the learning and control capabilities of the proposed RNN by providing illustrative simulations and experiments on a Franka Emika Panda robot. The results exhibit significant collision avoidance and learning performance of the proposed method with tiny errors.

关键词： Robots Collision avoidance End effectors Mathematical models Kinematics Jacobian matrices Service robots Vectors Robot sensing systems Sun Neural dynamics method obstacle avoidance recurrent neural network (RNN) redundant robots

来源：评论

学校读者我要写书评

暂无评论

High-Confidence data-driven Safe Tracking control Design

引用

ieee TRANSACTIONS ON CYBERNETICS 2025年第6期55卷 2583-2596页

作者： Niknejad, Nariman Esmzad, Ramin Modares, Hamidreza Michigan State Univ Dept Mech Engn E Lansing MI 48863 USA

This article presents a high-confidence data-driven safe tracking control design for stochastic linear discrete-time systems. The high-confidence safe reference tracking for an ellipsoidal safe set is first formalized using the concept of probabilistic set-based lambda-contractivity. A data-driven controller, composed of feedback and feedforward elements, is then designed to enforce the lambda-contractivity of the safe set. The feedback control gain is learned by 1) providing a data-driven representation of the closed-loop system, which contains a decision variable that affects the control gain and 2) optimizing the decision variable to ensure the lambda-contractivity. This feedback term can be learned using a data set that is not even rich enough to identify the full system model. A feedforward gain learning algorithm and a data-driven reference governor are provided to satisfy the required conditions on equilibrium terms. It is shown that under certain conditions on the equilibrium terms, the learned tracking controller guarantees the system's safety and stability with high probability. The reference governor dynamically manipulates the desired reference signal based on the data quality to prevent any breach of safety constraints in a probabilistic manner. It is shown that the output of the reference governor eventually converges to the desired goal states if inside the safe set and high-quality data is available. Therefore, the tracking controller guarantees convergence of the system output to its desired goal while ensuring safety with a high probability. The simulation results on a drone hovering and a test system, comparing the results with the existing literature, confirm that the presented high-confidence data-driven safe tracking control outperforms certainty-equivalent safe control methods.

关键词： Safety Probabilistic logic Noise control systems Uncertainty Stochastic systems Feedforward systems control design Symmetric matrices Stochastic processes data-driven control linear matrix inequalitie (LMIs) stochastic control tracking control

来源：评论

学校读者我要写书评

暂无评论

Bayesian Formal Synthesis of Unknown systems via Robust Simulation Relations

引用

ieee TRANSACTIONS ON AUTOMATIC control 2025年第3期70卷 1531-1546页

作者： Schoen, Oliver van Huijgevoort, Birgit Haesaert, Sofie Soudjani, Sadegh Newcastle Univ Sch Comp Newcastle Upon Tyne NE4 5TG England Eindhoven Univ Technol TU Elect Engn Dept NL-5612 Eindhoven Netherlands Max Planck Inst Software Syst D-67663 Kaiserslautern Germany

This article addresses the problem of data-driven computation of controllers that are correct by design for safety-critical systems and can provably satisfy (complex) functional requirements. With a focus on continuous-space stochastic systems with parametric uncertainty, we propose a two-stage approach that decomposes the problem into a learning stage and a robust formal controller synthesis stage. The first stage utilizes available Bayesian regression results to compute robust credible sets for the true parameters of the system. For the second stage, we introduce methods for systems subject to both stochastic and parametric uncertainties. We provide simulation relations for enabling correct-by-design control refinement that are founded on coupling uncertainties of stochastic systems via subprobability measures. The presented relations are essential for constructing abstract models that are related to not only one model but to a set of parameterized models. The results are demonstrated on three case studies, including a nonlinear and a high-dimensional system.

关键词： Uncertainty control systems Stochastic systems Stochastic processes data models Bayes methods Adaptation models Abstractions Bayesian regression data-driven methods incomplete knowledge simulation relations stochastic systems system identification temporal logic control

来源：评论

学校读者我要写书评

暂无评论

Safe data-driven Predictive control

引用

ieee Transactions on Automation Science and Engineering 2025年 22卷 14422-14435页

作者： Vahidi-Moghaddam, Amin Chen, Kaian Zhang, Kaixiang Li, Zhaojian Wang, Yan Wu, Kai Michigan State University Department of Mechanical Engineering East LansingMI48824 United States Ford Motor Company Research and Advanced Engineering DearbornMI48121 United States

In the realm of control systems, model predictive control (MPC) has exhibited remarkable potential;however, its reliance on accurate models and substantial computational resources has hindered its broader application, especially within real-time nonlinear systems. This study presents an innovative control framework to enhance the practical viability of the MPC. The developed safe data-driven predictive control aims to eliminate the requirement for precise models and alleviate computational burdens in the nonlinear MPC (NMPC). This is achieved by learning both the system dynamics and the control policy, enabling efficient data-driven predictive control while ensuring system safety. The methodology involves a spatial temporal filter (STF)-based concurrent learning for system identification, a robust control barrier function (RCBF) to ensure the system safety amid model uncertainties, and a RCBF-based NMPC policy approximation. An online policy correction mechanism is also introduced to counteract performance degradation caused by the existing model uncertainties. Demonstrated through simulations on two applications, the proposed approach offers comparable performance to existing benchmarks with significantly reduced computational costs. © 2004-2012 ieee.

关键词： Predictive control Uncertainty Safety Computational modeling System identification Adaptation models Trajectory System dynamics Robust control Predictive models

来源：评论

学校读者我要写书评

暂无评论

data-driven Combined Longitudinal and Lateral control for the Car Following Problem

引用

ieee TRANSACTIONS ON control systems TECHNOLOGY 2025年第3期33卷 991-1005页

作者： Cui, Leilei Chakraborty, Sayan Ozbay, Kaan Jiang, Zhong-Ping MIT Cambridge MA 02139 USA NYU Tandon Sch Engn Dept Elect & Comp Engn Control & Networks Lab Brooklyn NY 11201 USA NYU C2SMARTER Ctr Tandon Sch Engn Dept Civil & Urban Engn Brooklyn NY 11201 USA NYU Dept Elect & Comp Engn Dept Civil & Urban Engn Control & Networks LabTandon Sch Engn Brooklyn NY 11201 USA

This article studies the problem of data-driven combined longitudinal and lateral control of autonomous vehicles (AVs) such that the AV can stay within a safe but minimum distance from its leading vehicle and, at the same time, in the lane. Most of the existing methods for combined longitudinal and lateral control are either model-based or developed by purely data-driven methods such as reinforcement learning. Traditional model-based control approaches are insufficient to address the adaptive optimal control design issue for AVs in dynamically changing environments and are subject to model uncertainty. Moreover, the conventional reinforcement learning approaches require a large volume of data, and cannot guarantee the stability of the vehicle. These limitations are addressed by integrating the advanced control theory with reinforcement learning techniques. To be more specific, by utilizing adaptive dynamic programming (ADP) techniques and using the motion data collected from the vehicles, a policy iteration algorithm is proposed such that the control policy is iteratively optimized in the absence of the precise knowledge of the AV's dynamical model. Furthermore, the stability of the AV is guaranteed with the control policy generated at each iteration of the algorithm. The efficiency of the proposed approach is validated by the integrated simulation of SUMO and CommonRoad.

关键词： Adaptation models Vehicle dynamics Mathematical models Transportation Roads Reinforcement learning Nonlinear dynamical systems Electronic mail Dynamic programming Accuracy Adaptive dynamic programming (ADP) combined longitudinal and lateral control connected vehicles

来源：评论

学校读者我要写书评

暂无评论

The Q-Fractionalism Reasoning learning Method

引用

ieee TRANSACTIONS ON NEURAL NETWORKS AND learning systems 2025年第1期36卷 1568-1582页

作者： Mazandarani, Mehran Jianfei, Pan Shenzhen Univ Dept Mechatron & Control Engn Shenzhen 518060 Peoples R China

As the title suggests, in this work, a modern machine learning method called the Q-fractionalism reasoning is introduced. The proposed method is founded upon a synergy of the Q-learning and fractional fuzzy inference systems (FFISs). Unlike other approaches, the Q-fractionalism reasoning not only incorporates the knowledge base to understand how to perform but also explores a reasoning mechanism from the fractional order to justify what it has performed. This method suggests that the agent choose actions aimed at the characterization of reasoning. In fact, the agent deals with states termed as primary and secondary fuzzy states. The primary fuzzy states are unobservable and uncertain, for which the agent chooses actions. However, the projection of primary fuzzy states onto the knowledge base results in secondary fuzzy states, which are observable by the agent, allowing it to detect primary fuzzy states with degrees of detectability. With a practical experiment implemented on a linear switched reluctance motor (LSRM), the results demonstrate that the application of the Q-fractionalism reasoning in the real-time position control of the LSRM leads to a remarkable improvement of about 70% in the accuracy of the control objective compared with a typical fuzzy inference system (FIS) under the same setting.

关键词： Artificial intelligence computational intelligence data-driven control systems fuzzy systems reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Applications of data-driven Dynamic Modeling of Power Converters in Power systems: An Overview

引用

ieee TRANSACTIONS ON INDUSTRY APPLICATIONS 2025年第2期61卷 2434-2456页

作者： Subedi, Sunil Gui, Yonghao Xue, Yaosuo Oak Ridge Natl Lab Electrificat & Energy Infrastruct Div Oak Ridge TN 37831 USA

Power electronic converter (PEC)-based resources are growing ubiquitously in power systems and there is a vital necessity for precise dynamic models to comprehend their dynamics to different events and control strategies. Inaccurate modeling can lead to instability, higher costs, and reliability issues. Anticipating the increase in PECs in the near future, detailed modeling becomes computationally and mathematically complex, requiring extensive computing power and knowledge of vendor-specific PECs. To overcome these challenges, data-driven machine learning/artificial intelligence (ML/AI) approaches are widely used, tracking the dynamic responses of PECs operating in various modes with limited knowledge. These models find applications in protection, stability, fault diagnosis, optimization, control and monitoring, and power quality. While the literature on power systems often emphasizes the advantages of data-driven modeling, an in-depth look at the limitations, challenges, and opportunities related to converter-dominated grids is still lacking. The purpose of this survey is to conduct a comprehensive review of ML/AI methodologies in PECs and investigate their applications in power systems. The article introduces various PEC types, their roles, and modeling approaches. It then provides an in-depth overview of how ML/AI can be applied to PECs in power systems. Finally, the survey highlights gaps in the field's knowledge and suggests potential directions for future research.

关键词： Power system stability Power system dynamics Computational modeling Reviews Hidden Markov models Vehicle dynamics Topology Load modeling Matrix converters Dynamic scheduling Artificial intelligence converter-dominated data-driven modeling dynamics fault diagnosis power electronic converter power quality protection stability

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

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

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：