检索结果-内蒙古大学图书馆

Robust Fault-tolerant Tracking Control for Linear Discrete-time Systems via reinforcement learning method

INTERNATIONAL JOURNAL OF CONTROL AUTOMATION AND SYSTEMS 2025年第2期23卷 520-529页

作者： Nguyen, Ngoc Hoai An Kim, Sung Hyun Univ Ulsan Dept Elect Elect & Comp Engn Ulsan 680749 South Korea

Concentrated on the off-policy reinforcement learning method, this paper explores a model-free algorithm for addressing the robust fault-tolerant tracking problem in discrete-time linear systems with time-varying actuator faults and model uncertainties. Specifically, to determine the feedback control input, a dynamic optimization approach is developed based on measured data rather than exact information from system dynamics. Subsequently, a static optimization approach is established using solutions from the preceding dynamic optimization problem to compute the feedforward control input. Finally, numerical simulations are conducted to illustrate the feasibility and efficiency of the proposed solution.

关键词： Actuator faults fault-tolerant control model uncertainties off-policy approach reinforcement learning method robust control tracking control

来源：评论

学校读者我要写书评

暂无评论

reinforcement learning method for QoE-aware Optimization of Content Delivery

Reinforcement Learning Method for QoE-aware Optimization of ...

引用

IEEE Wireless Communications and Networking Conference (WCNC)

作者： Yousaf, Faqir Zarrar Mammela, Olli Mannersalo, Petteri NEC Labs Europe Heidelberg Germany VTT Tech Res Ctr Finland Espoo Finland

ISBN: (纸本)9781479930838

The delivery of video services in a controllable and resource efficient manner while meeting the various QoE/QoS requirements in mobile networks is a challenging task, especially in a multiclass wireless environment. This paper proposes an intelligent and context-aware application level fair scheduler based on reinforcement-learning, which can dynamically adjust relevant scheduling parameters in reaction to specific events or context information. The implemented Q-learning method is analyzed with reference to the delivery of progressive video streaming services. We first highlight the performance issues during progressive video streaming over TCP to multiple users under resource constrained environment. We then demonstrate the utilization of employing Q-learning method in our scheduler for intelligent orchestration between multiple concurrent flows to ensure against buffer starvation and thus enable smooth playback. We also demonstrate the effectiveness of our context aware dynamic scheduler to provide service separation between the user classes and fairness within a user class.

关键词： dynamic scheduling learning (artificial intelligence) mobile communication mobile computing optimisation quality of experience quality of service transport protocols video streaming Q-learning method QoS TCP protocol content delivery QoE-aware optimization context-aware dynamic scheduler intelligent orchestration mobile network multiclass wireless environment multiple concurrent flow reinforcement learning method resource constrained environment scheduling parameter video streaming service delivery Analytical models Bandwidth Delays Gold Mobile communication Silver Streaming media mobile communication Quality of service Video streaming mobile networking Transmission control protocol Streaming Video learning methods dynamic scheduling transport protocols Mobile Computing ANALYSIS MODELS Bandwidth Content distribution Silver Gold

来源：评论

学校读者我要写书评

暂无评论

reinforcement learning method-based stable gait synthesis for biped robot

Reinforcement learning method-based stable gait synthesis fo...

引用

8th International Conference on Control, Automation, Robotics and Vision (ICARCV 2004)

作者： Hu, LY Sun, ZQ Tsinghua Univ Comp Sci & Technol Dept State Key Lab Intelligent Technol & Syst Beijing 100084 Peoples R China

ISBN: (纸本)0780386531

A stable gait generation algorithm based on T-S type fuzzy learning net is proposed in this paper. Gait generation is divided into model construction and error learning. Reference gait model and dynamic model are firstly constructed with basic gait geometric knowledge. Then reinforcement learning method is introduced into T-S type fuzzy network to learn the gain parameters for hip trajectory adjustment. Few fuzzy rules with ZMP stable knowledge are needed to formulate the nonlinear relation between the ZMP curve and hip trajectory. The problem of finding multi-variables in continuous space is also simplified to searching independent action gains simultaneously. Results of simulation on a biped robot proved the feasibility.

关键词： learning (artificial intelligence) gait analysis legged locomotion fuzzy neural nets robot dynamics reinforcement learning method stable gait synthesis biped robot stable gait generation algorithm T-S type fuzzy learning network model construction error learning dynamic model gait geometric knowledge ZMP stable knowledge

来源：评论

学校读者我要写书评

暂无评论

Swarm reinforcement learning method Based on Hierarchical Q-learning

Swarm Reinforcement Learning Method Based on Hierarchical Q-...

引用

IEEE Symposium Series on Computational Intelligence (IEEE SSCI)

作者： Kuroe, Yasuaki Takeuchi, Kenya Maeda, Yutaka Kansai Univ Fac Engn Sci Suita Osaka Japan

ISBN: (纸本)9781728190488

In last decades the reinforcement learning method has attracted a great deal of attention and many studies have been done. However, this method is basically a trial-and-error scheme and it takes much computational time to acquire optimal strategies. Furthermore, optimal strategies may not be obtained for large and complicated problems with many states. To resolve these problems we have proposed the swarm reinforcement learning method, which is developed inspired by the multi-point search optimization methods. The Swarm reinforcement learning method has been extensively studied and its effectiveness has been confirmed for several problems, especially for Markov decision processes where the agents can fully observe the states of environments. In many real-world problems, however, the agents cannot fully observe the environments and they are usually partially observable Markov decision processes (POMDPs). The purpose of this paper is to develop a swarm reinforcement learning method which can deal with POMDPs. We propose a swarm reinforcement learning method based on HQ-learning, which is a hierarchical extension of Q-learning. It is shown through experiments that the proposed method can handle POMDPs and possesses higher performance than that of the original HQ-learning.

关键词： reinforcement learning method partially observed Markov decision process hierarchical Q-learning swarm intelligence

来源：评论

学校读者我要写书评

暂无评论

An electrical vehicle-assisted demand response management system: A reinforcement learning method

引用

FRONTIERS IN ENERGY RESEARCH 2023年 10卷

作者： Li, Donghe Yang, Qingyu Ma, Linyue Wang, Yiran Zhang, Yang Liao, Xiao Xi An Jiao Tong Univ Sch Automat Sci & Engn Xian Shaanxi Peoples R China Xi An Jiao Tong Univ State Key Lab Mfg Syst Engn Xian Shaanxi Peoples R China State Grid Informat & Telecommun Grp Co LTD Beijing Peoples R China

With the continuous progress of urbanization, determining the charging and discharging strategy for randomly parked electric vehicles to help the peak load shifting without affecting users' travel is a key problem. This paper design a reinforcement learning-based method for the electric vehicle-assisted demand response management system. Specifically, we formalize the charging and discharging sequential decision problem of the parking lot into the Markov process, in which the state space is composed of the state of parking spaces, electric vehicles, and the total load. The charging and discharging decision of each parking space acts as the action space. The reward comprises the penalty term that guarantees the user's travel and the sliding average value of the load representing peak load shifting. After that, we use a Deep Q-Network (DQN)-based reinforcement learning architecture to solve this problem. Finally, we conduct a comprehensive evaluation with real-world power usage data. The results show that our proposed method will reduce the peak load by 10% without affecting the travel plan of all electric vehicles. Compared with random charging and discharging scenarios, we have better performance in terms of state-of-charge (SoC) achievement rate and peak load shifting effect.

关键词： demand response electric vehicle reinforcement learning method MDP DQN

来源：评论

学校读者我要写书评

暂无评论

End-to-End Deep Policy Feedback-Based reinforcement learning method for Quantization in DNNs

引用

JOURNAL OF CIRCUITS SYSTEMS AND COMPUTERS 2022年第13期31卷

作者： Babu, R. Logesh Gurumoorthy, Sasikumar Parameshachari, B. D. Nelson, S. Christalin Hua, Qiaozhi Madanapalle Inst Technol & Sci Dept Comp Sci & Engn Chittoor 517325 Andhra Pradesh India Jerusalem Coll Engn Dept Comp Sci & Engn Chennai 600100 Tamil Nadu India GSSS Inst Engn & Technol Women Dept Telecommun Engn Mysuru 570011 Karnataka India Univ Petr & Energy Studies UPES Sch Comp Sci Dept Syst Cluster Dehra Dun 248007 Uttarakhand India Hubei Univ Arts & Sci Sch Comp Xiangyang 441000 Hubei Peoples R China

In the resource-constrained embedded systems, the designing of efficient deep neural networks is a challenging process, due to diversity in the artificial intelligence applications. The quantization in deep neural networks superiorly diminishes the storage and computational time by reducing the bit-width of networks encoding. In order to highlight the problem of accuracy loss, the quantization levels are automatically discovered using Policy Feedback-based reinforcement learning method (PF-RELEQ). In this paper, the Proximal Policy Optimization with Policy Feedback (PPO-PF) technique is proposed to determine the best design decisions by choosing the optimum hyper-parameters. In order to enhance the sensitivity of the value function to the change of policy and to improve the accuracy of value estimation at the early learning stage, a policy update method is devised based on the clipped discount factor. In addition, specifically the loss functions of policy satisfy the unbiased estimation of the trust region. The proposed PF-RELEQ effectively balances quality and speed compared to other deep learning methods like ResNet-1202, ResNet-32, ResNet-110, GoogLeNet and AlexNet. The experimental analysis showed that PF-RELEQ achieved 20% computational work-load reduction compared to the existing deep learning methods on ImageNet, CIFAR-10, CIFAR-100 and tomato leaf disease datasets and achieved approximately 2% of improvisation in the validation accuracy. Additionally, the PF-RELEQ needs only 0.55 Graphics Processing Unit on an NVIDIA GTX-1080Ti to develop DNNs that delivers better accuracy improvement with fewer cycle counts for image classification.

关键词： Constrained embedded systems deep neural networks long short-term memory network policy feedback proximal policy optimization technique reinforcement learning method

来源：评论

学校读者我要写书评

暂无评论

Solving Network Design Problem with Dynamic Network Loading Profiles Using Modified reinforcement learning method

引用

Procedia - Social and Behavioral Sciences 2014年 111卷 38-47页

作者： Cenk Ozan Halim Ceylan Soner Haldenbilen Department of Civil Engineering Pamukkale University Kinikli Campus Denizli 20070 Turkey

This study aims to solve dynamic user Equilibrium Network Design Problem (ENDP) with dynamic network loading profiles using modified reinforcement learning (RL) approach. The bi-level programming technique is used to solve the problem. At the lower level of the problem, the dynamic User Equilibrium (UE) link flows are obtained by simulation based Dynamic Traffic Assignment (DTA) model with DynusT and signal timings are obtained at the upper level by modified RL method. The system Performance Index (PI) is defined as the sum of a weighted linear combination of delay and number of stops per unit time for all traffic streams, which is evaluated by the traffic model of TRANSYT-7F. Q-learning , a model-free approach, is one of the RL methods. The modified RL method is actually based on Q-learning . By integrating the modified RL method, traffic assignment and traffic control, the modified RE inforcement L earning TRA NSYT-7F D ynusT (RELTRAD) model is proposed to solve the dynamic ENDP. The objective function of the proposed RELTRAD is total network PI. The model is tested on the medium sized Allsop and Charlesworth's network. Two scenarios, related to various dynamic network loading profiles, are proposed for numerical application. Encouraging results are obtained. Results showed that the RELTRAD model effectively optimizes the signal timings and values of the network PI. The RELTRAD model improves to the network PI from the initial value to the final value as 65% and 67% for loading profile 1 and 2, respectively.

关键词： Network design problem reinforcement learning method dynamic network loading DynusT TRANSYT-7F

来源：评论

学校读者我要写书评

暂无评论

Design of ABR Flow Controller Based on reinforcement learning-PID method

Design of ABR Flow Controller Based on Reinforcement Learnin...

引用

20th Chinese Control and Decision Conference

作者： Zhao, Xin Li, Xin Shenyang Inst Aeronaut Engn Sch Elect & Informat Engn Shenyang 110136 Peoples R China Northeastern Univ Coll Informat Sci & Engn Shenyang 110004 Peoples R China

ISBN: (纸本)9781424417339

For the congestion problems in asynchronous transfer mode (ATM) networks, the reinforcement learning method is used in the updating of the control parameters, and a controller based on reinforcement learning-PID method is proposed. This algorithm is independent of the mathematic model and priori-knowledge of networks. It obtains the knowledge through trial-and-error and interaction with the environment to improve its behavior policy So it has the ability of self-learning. The controller forces the queue length at the bottleneck node to the desired value by adjusting the source traffic rate of the available bit rate (ABR) service and guarantees the stability of the system to avoid the occur of congestion. Simulation results show the effectiveness of the method.

关键词： ABR service flow control reinforcement learning method PID control

来源：评论

学校读者我要写书评

暂无评论

Formulating and analysis of traffic flow to secure software-defined network (SDN) using recursive network (RN) learning method

引用

JOURNAL OF SUPERCOMPUTING 2025年第5期81卷

作者： Ram, Anil Chakraborty, Swarnendu Kumar Banerjee, Aiswarrya Mahato, Ganesh Kumar Natl Inst Technol Arunachal Pradesh Dept Comp Sci & Engn Jote 791113 Arunachal Prade India

This work proposes a new solution for monitoring traffic flow (TFM) in the context of software-defined network (SDN) systems, particularly focusing on the detection of cyberattacks like DDoS. The solution is based on a Recursive Network (RN) design which enables higher accuracy in an ensemble of anomaly detection, which ensures complete SDN data plane protection even during the conditions of attack. The goal of this solution is to retain the highest possible network forwarding efficiency and avoid network congestion during cyberthreats. This approach mitigates overwhelming DDoS attack attempts characteristic to SDNs because of their centralized and ever-changing design. More conventional ways of detection tend to be blind to such attacks-this is why their detection becomes detection innovative. Gradually checking the traffic flow through the RN structure allows for more precise targeting of abnormal patterns often associated with misconduct behaviors. A key construct for this methodology is the implementation of a learned policy to control dynamic transit flow. The policy is aimed at optimizing the collection of flow data at points of data acquisition in order to provide prompt assessment of superfluous conditions, and therefore a prompt assessment to possible threats. The derived flow matching parameters respond dynamically to formed traffic patterns, and as a result the system gains additional abilities to avert cyberattacks. Comparative evaluations demonstrate the efficiency of the proposed model, with performance improvements ranging from 15.29 to 48.13% over existing methods, such as FMS, MMOS, DATA, QDATA, and DEEP-MC. This showcases the proposed method's superiority in enhancing cyberattack detection within SDN environments. By preventing network degradation during DDoS attacks and other anomalies, this solution ensures the integrity and availability of the SDN's data plane, contributing to the protection of critical network infrastructure.

关键词： SDN RN learning Traffic flow monitoring reinforcement learning method Security Cyber attack detection

来源：评论

学校读者我要写书评

暂无评论

Comparison of Model Predictive and reinforcement learning methods for Fault Tolerant Control

引用

IFAC-PapersOnLine 2018年第24期51卷 233-240页

作者： Ahmed, Ibrahim Khorasgani, Hamed Biswas, Gautam Vanderbilt University United States Institute of Software Integrated Systems Vanderbilt University United States

A desirable property in fault-tolerant controllers is adaptability to system changes as they evolve during systems operations. An adaptive controller does not require optimal control policies to be enumerated for possible faults. Instead it can approximate one in real-time. We present two adaptive fault-tolerant control schemes for a discrete time system based on hierarchical reinforcement learning. We compare their performance against a model predictive controller in presence of sensor noise and persistent faults. The controllers are tested on a fuel tank model of a C-130 plane. Our experiments demonstrate that reinforcement learning-based controllers perform more robustly than model predictive controllers under faults, partially observable system models, and varying sensor noise levels. © 2018

关键词： Adaptive control systems Controllers Digital control systems Discrete time control systems Fault tolerance Fuel tanks Model predictive control reinforcement learning Adaptive fault tolerant control Fault tolerant controllers Hierarchical reinforcement learning Model based controls Model predictive controllers Partially observable systems reinforcement learning control reinforcement learning method

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：