检索结果-内蒙古大学图书馆

International IEEE Conference on Intelligent Transportation Systems

作者： Junchen Jin Xiaoliang Ma Div. of Transp. Planning Econ. & Eng. KTH R. Inst. of Technol. Stockholm Sweden

ISBN: (纸本)9781467365970

Group-based signal controllers are widely deployed on urban networks in the European countries. However, group-based signal controls are usually implemented with rather simple timing logics, e.g. vehicle actuated timing. In addition, group-based signal control systems with pre-defined signal parameter settings show relatively poor performances in a dynamically changed traffic environment. This study, therefore, presents an adaptive group-based signal control system capable of changing control strategies with respect to non-stationary traffic demands. In this study, signal groups are formulated as individual agents. The signal group agent learns from traffic environments and makes intelligent timing decisions according to the perceived system states. reinforcement learning with multiple-step backups is applied as the learning algorithm. Agents on-line update their knowledge based on a sequence of states during the learning process rather than purely on the basis of single previous state. The proposed signal control system is integrated into a software-in-the-loop simulation (SILS) framework for evaluation purpose. In the testbed experiments, the proposed adaptive group-based control system is compared to a benchmark signal control system, the well-established group-based fixed-time control system. The simulation results demonstrate that learning-based and adaptive group-based signal control system owns its advantage in dealing with dynamic traffic environments in terms of improving traffic mobility efficiency.

关键词： adaptive control control engineering computing digital simulation learning (artificial intelligence) road traffic control traffic engineering computing Nordic countries SILS framework adaptive group-based signal control system agent learning group-based fixed-time control system reinforcement learning algorithm signal parameter settings software-in-the-loop simulation timing logic traffic mobility efficiency urban networks Adaptive systems Control systems Delays Green products learning (artificial intelligence) Vehicles

来源：评论

学校读者我要写书评

暂无评论

Using hybrid multiobjective machine learning to optimise sonobuoy placement patterns

引用

IET RADAR SONAR AND NAVIGATION 2023年第3期17卷 374-387页

作者： Taylor, Christopher M. Maskell, Simon Ralph, Jason F. Univ Liverpool Dept Elect & Elect Engn Brownlow Hill Liverpool L69 3GJ Merseyside England

This paper presents a new approach to finding optimal patterns for the placement of fields of sonobuoys in a complex undersea environment. We model the problem as a biobjective one, where the aim is to minimise both sensor placement time and uncertainty over target localisation. Both objectives may be important in time-critical localisation scenarios and our approach allows an operator to choose between different optimal solutions, favouring lower placement time or lower localisation uncertainty as operational circumstances require. We develop a two-phase algorithm, where an offline multiobjective evolutionary phase finds initial Pareto-non-dominated solutions to a static problem and then an online multiobjective reinforcement learning phase finds improved solutions using updated information. We find that the evolutionary algorithm improves significantly on standard grid patterns and that the reinforcement learning algorithm improves further on the evolutionary phase. The number of sonobuoys required may also be reduced.

关键词： optimal patterns Optimisation techniques static problem standard grid patterns Pareto optimisation sensor placement time evolutionary computation offline multiobjective evolutionary phase telecommunication computing initial Pareto-nondominated solutions operational circumstances complex undersea environment placement time localisation uncertainty reinforcement learning algorithm sonobuoy placement patterns evolutionary algorithm online multiobjective reinforcement learning phase two-phase algorithm reinforcement learning target localisation time-critical localisation scenarios hybrid multiobjective machine Communications computing reinforcement learning sensor placement Wireless sensor networks

来源：评论

学校读者我要写书评

暂无评论

Optimizing QoS routing in hierarchical ATM networks using computational intelligence techniques

引用

IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART C-APPLICATIONS AND REVIEWS 2003年第3期33卷 297-312页

作者： Vasilakos, A Saltouros, MP Atlassis, AF Pedrycz, W FORTH Fdn Res & Technol Hellas Inst Comp Sci Iraklion 15410 Greece Natl Tech Univ Athens Dept Elect & Comp Engn GR-15773 Athens Greece Univ Alberta Dept Elect & Comp Engn Edmonton AB T6G 2G7 Canada

In this paper, the use of a computational intelligence approach -a reinforcement learning algorithm (RLA)-for optimizing the routing in asynchronous transfer mode (ATM) networks based on the private network-to-network interface (PNNI) standard is proposed. This algorithm which is specially designed for the quality of service (QoS) routing problem, aims at maximizing the network revenue (allocating efficiently the network resources) while ensuring the QoS requirements for each connection. In this study, large-scale networks are considered where it becomes necessary to be organized hierarchically so that a scale in terms of computation, communication and storage requirements will be achieved. A comparative performance study of the proposed and other commonly used routing schemes is demonstrated by means of simulation on existing commercial networks. Simulation results over a wide range of uniform, time-varying and skewed loading conditions show the effectiveness of the proposed routing algorithm, and disclose the strength and weakness of the various schemes.

关键词： ATM QoS routing computational intelligence reinforcement learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Research on coordinated scheduling of straddle carriers and quay cranes in automated container terminals based on reinforcement learning

Research on coordinated scheduling of straddle carriers and ...

引用

作者： Zhenyu Fang Xiaolong Han Shanghai Maritime University Institute of Logistics Science & Engineering

Aiming at the coordination and scheduling problem between Automated Straddle Carrier(Automated Straddle Carrier) and Quay Crane(QC) in automated container terminals,consider that the quay crane cannot cross the straddle carrier,and the straddle carrier and the quay crane cannot be in adjacent lanes at the same time.A mixed-integer optimization model with the objective function of minimizing the final completion time of the quay crane is established under the constraints of different operating speeds and different speeds of straddle carriers in different states.A genetic algorithm based on reinforcement learning is designed,and the initial population is generated by the Q-learning algorithm,and the genetic algorithm(GA) is iterated to increase the diversity of the initial ***,taking 5 groups of experiments as examples,the model is compared and solved by GAMS solver,genetic algorithm and genetic algorithm based on reinforcement *** results of an example show that the genetic algorithm based on reinforcement learning can solve the model and make the value of the objective function smaller,thus verifying the feasibility of the modified algorithm.

关键词： Automated container terminal straddle carrier quay crane coordinate scheduling reinforcement learning algorithm genetic algorithm

来源：评论

学校读者我要写书评

暂无评论

Data-driven optimal control of operational indices for a class of industrial processes

引用

IET CONTROL THEORY AND APPLICATIONS 2016年第12期10卷 1348-1356页

作者： Lu, Xinglong Kiumarsi, Bahare Chai, Tianyou Lewis, Frank L. Northeastern Univ State Key Lab Synthet Automat Proc Ind Shenyang 110819 Peoples R China Univ Texas Arlington UTA Res Inst Ft Worth TX 76118 USA

In this study, a data-driven optimisation solution for operational index control for a class of industrial processes is presented. First, the operational index control problem is formulated as an optimal tracking control problem. Then, an augmented system composed of the device loop dynamics and operational indices dynamics is constructed on two different time scales. Since, finding mathematical model of the operational indices dynamics is difficult, in contrast to most existing operational optimisation and control methods that use a mathematical model of the operational indices dynamics, a reinforcement learning algorithm based on actor-critic structure is employed to provide a data-driven optimisation control method to select optimal process setpoints so that the operational indices can track desired values. This solution does not require complete knowledge of the industrial process dynamics. Moreover, complicated system identification of the dynamics of the operational indices is not required. The effectiveness of the proposed method is demonstrated by experimental results that are carried out on a hardware-in-the-loop emulation system for a mineral grinding process.

关键词： process control optimal control optimisation learning (artificial intelligence) mineral processing grinding control engineering computing data-driven optimal control data-driven optimisation solution operational index control problem optimal tracking control problem augmented system device loop dynamics operational index dynamics mathematical model reinforcement learning algorithm actor-critic structure optimal process setpoint selection industrial process dynamics mineral grinding process hardware-in-the-loop emulation system

来源：评论

学校读者我要写书评

暂无评论

Online two-timescale service placement for time-sensitive applications in MEC-assisted network: A TMAGRL approach

引用

COMPUTER NETWORKS 2024年 244卷

作者： Du, An Jia, Jie Chen, Jian Guo, Liang Wang, Xingwei Northeastern Univ Sch Comp Sci & Engn Shenyang 110819 Liaoning Peoples R China Minist Educ Engn Res Ctr Secur Technol Complex Network Syst Shenyang 110819 Liaoning Peoples R China Northeastern Univ Key Lab Intelligent Comp Med Image Minist Educ Shenyang 110819 Liaoning Peoples R China

Mobile edge computing (MEC) integrated with the Network Functions Virtualization (NFV) technique has been regarded as a promising solution for flexible services provision and user service experience improvement. However, existing service placement in such systems still faces the challenge of satisfying computing tasks with strict latency requirements, especially when massive mobile users roam around different coverage areas of edge servers. For this purpose, we first adopt a novel service placement framework that combines proactive replicas pre -deployment and reactive service migration. Based on this, we investigate the dynamic placement problem of multiple types of services achieved by the various virtualized network functions (VNFs) to minimize long-term redeployment costs in MEC -assisted systems, subject to the completion deadline of tasks and limited computing resources of edge servers. Considering that the update timescale of VNF replicas pre -deployment is different, we design a novel two-timescale multi -agent graph convolutional network -based reinforcement learning algorithm (TMAGRL) by invoking a long-timescale training layer for proactive VNF replicas placement and a short-timescale training layer for reactive VNF migration. Extensive numerical results reveal that TMAGRL, based on the designed hybrid framework, can learn a VNF placement strategy to adapt to the dynamics of the system without any prior information. Moreover, we verify its superior performance in terms of average service response latency and overall redeployment cost by comparing it with baselines.

关键词： Mobile edge computing Proactive replicas pre-deployment Reactive service migration reinforcement learning algorithm Time-sensitive

来源：评论

学校读者我要写书评

暂无评论

Using a Collaborative Robot to the Upper Limb Rehabilitation 4th

Using a Collaborative Robot to the Upper Limb Rehabilitation

引用

4th Iberian Robotics Conference (Robot) - Advances in Robotics

作者： Fernandes, Lucas de Azevedo Lima, Jose Luis Leitao, Paulo Nakano, Alberto Yoshiro Univ Tecnol Fed Parana Curitiba Parana Brazil Polytech Inst Braganca CeDRI Res Ctr Digitalizat & Intelligent Robot Porto Portugal INESC TEC Porto Portugal

ISBN: (纸本)9783030361501;9783030361495

Rehabilitation is a relevant process for the recovery from dysfunctions and improves the realization of patient's Activities of Daily Living (ADLs). Robotic systems are considered an important field within the development of physical rehabilitation, thus allowing the collection of several data, besides performing exercises with intensity and repeatedly. This paper addresses the use of a collaborative robot applied in the rehabilitation field to help the physiotherapy of upper limb of patients, specifically shoulder. To perform the movements with any patient the system must learn to behave to each of them. In this sense, the reinforcement learning (RL) algorithm makes the system robust and independent of the path of motion. To test this approach, it is proposed a simulation with a UR3 robot implemented in V-REP platform. The main control variable is the resistance force that the robot is able to do against the movement performed by the human arm.

关键词： Robotics rehabilitation Collaborative robots Simulation reinforcement learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Multiagent-based Market Simulator for the Wholesale Electricity Spot Market

Multiagent-based Market Simulator for the Wholesale Electric...

引用

IEEE Region 10 Conference (TENCON) - Sustainable Development through Humanitarian Technology

作者： Pacaba, Dominic Dave P. Nerves, Allan C. Univ Philippines Diliman Elect & Elect Engn Inst Quezon City Philippines

ISBN: (纸本)9781467348232

As electricity markets develop into more complex structures, new modeling and simulation techniques are required to simulate the market and to identify strategic behavior that can profitably influence electricity prices. This study proposes an agent-based model to characterize and investigate the complex interaction between the physical network infrastructure of the power system and the economic behavior of the electricity market. Market clearing prices are determined by a single-sided auction where the profit-seeking behavior of a generation company is modeled using an autonomous adaptive agent capable of endogenously developing its own bidding strategies through a Roth-Erev reinforcement learning algorithm that determine their price and quantity offers in the hourly auction. The agent-based model in this study implements the generator-reported cost function as a piecewise linear function in order to more accurately represent bidding behavior and strategies. Market clearing can then be formulated as a DC optimal power flow problem that is solved using linear programming. An agent-based model developed for the Wholesale Electricity Spot Market (WESM) in the Philippines is able to show the effects on locational marginal prices and network congestion and losses, of the profit-seeking and strategizing behavior of the generation companies.

关键词： agent-based modelling wholesale electricity markets reinforcement learning algorithm electricity market simulation

来源：评论

学校读者我要写书评

暂无评论

Simultaneous learning of Spatial Visual Attention and Physical Actions

Simultaneous Learning of Spatial Visual Attention and Physic...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems

作者： Borji, Ali Ahmadabadi, Majid Nili Araabi, Babak Nadjar Univ So Calif Dept Comp Sci Hedco Neurosci BldgRoom 93641 Watt Way Los Angeles CA 90089 USA Univ Tehran Sch Elect & Comp Engn Sch Cognit Sci Tehran Iran

ISBN: (纸本)9781424466757

This paper introduces a new method for learning top-down and task-driven visual attention control along with physical actions in interactive environments. Our method is based on the reinforcement learning of Visual Classes(RLVC) algorithm and adapts it for learning spatial visual selection in order to reduce computational complexity. Proposed algorithm also addresses aliasings due to not knowing previous actions and perceptions. Continuing learning shows our method is robust to perturbations in perceptual information. Our method also allows object recognition when class labels are used instead of physical actions. We have tried to gain maximum generalization while performing local processing. Experiments over visual navigation and object recognition tasks show that our method is more efficient in terms of computational complexity and is biologically more plausible.

关键词： computational complexity computational complexity generalisation (artificial intelligence) learning (artificial intelligence) navigation object recognition object recognition reinforcement learning algorithm spatial visual attention visual classes visual navigation

来源：评论

学校读者我要写书评

暂无评论

Using the Online Cross-Entropy Method to Learn Relational Policies for Playing Different Games

Using the Online Cross-Entropy Method to Learn Relational Po...

引用

IEEE Conference on Computational Intelligence and Games (CIG)

作者： Sarjant, Samuel Pfahringer, Bernhard Driessens, Kurt Smith, Tony Univ Waikato Fac Comp & Math Sci Dunedin New Zealand Maastricht Univ Dept Knowledge Engn NL-6200 MD Maastricht Netherlands

ISBN: (纸本)9781457700095

By defining a video-game environment as a collection of objects, relations, actions and rewards, the relational reinforcement learning algorithm presented in this paper generates and optimises a set of concise, human-readable relational rules for achieving maximal reward. Rule learning is achieved using a combination of incremental specialisation of rules and a modified online cross-entropy method, which dynamically adjusts the rate of learning as the agent progresses. The algorithm is tested on the Ms. Pac-Man and Mario environments, with results indicating the agent learns an effective policy for acting within each environment.

关键词： Computational intelligence Conferences Games Heuristic algorithms Junctions learning learning systems Mario environments Ms. Pac-Man computer games human readable relational rules learning (artificial intelligence) online cross entropy method playing different games reinforcement learning algorithm relational policies video game environment video signal processing

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：