检索结果-内蒙古大学图书馆

learning to construct a solution for UAV path planning problem with positioning error correction

KNOWLEDGE-BASED SYSTEMS 2024年 304卷

作者： Chun, Jie Chen, Ming Liu, Xiaolu Xiang, Shang Du, Yonghao Wu, Guohua Xing, Lining Natl Univ Def Technol Coll Syst Engn Changsha 410073 Hunan Peoples R China XiangTan Univ Sch Publ Adm Xiangtan 411100 Hunan Peoples R China Cent South Univ Sch Automat Changsha 410075 Hunan Peoples R China Xidian Univ Coll Elect Engn Xian 710126 Shanxi Peoples R China

Unmanned aerial vehicles (UAVs) are advanced flight systems. However, their positioning systems cause distance-dependent errors during flight. This study seeks to solve the UAV path planning problem with positioning error correction (UPEC) with an end-to-end method. Traditional methods struggle to balance solution quality and computational overload, and often have limited utilisation of scenario information. To overcome these issues, we propose a path planning model (PPM) based on deep reinforcement learning to solve the UPEC. The model has a complete structure that includes a mathematical model, feature engineering, solution process, neural policy network, scenario generation, training process, and test solution mechanism. Specifically, we first establish a Markov decision process (MDP) for UPEC and apply feature engineering with effective features to support decision-making. We then introduce a path planning neural network (PPNN) to represent the MDP policy. Based on the dataset generated from the multi-rule combination validation, we train the PPNN using the proposed RL algorithm with storage pool. Furthermore, we propose a backtracking mechanism to guarantee solution feasibility during the construction process. Extensive experiments demonstrate that the proposed PPM outperforms existing state-of-the-art algorithms in terms of solution quality and timeliness, and the backtracking mechanism effectively improves the scenario completion rate. The model study indicates the efficacy of our training algorithm and the generalisation of the PPNN. Additionally, our construction process is problem-tailored and more suitable for addressing UPEC than iterative search algorithms, because it effectively mitigates the impact of invalid nodes.

关键词： Deep reinforcement learning UAV Path planning Positioning error correction reinforcement learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Design of a Digital Exhibition Service System Under the Deep Belief Network Models

引用

IEEE ACCESS 2024年 12卷 108786-108796页

作者： Song, Qixin Tourism Coll Changchun Univ Sch Tourism & Culture Changchun 130607 Peoples R China Jilin Prov Res Ctr Cultural Tourism Educ & Enterp Changchun 130607 Peoples R China Northeast Asia Res Ctr Leisure Econ Changchun 130607 Peoples R China

This work aims to optimize the classification efficiency of the digital exhibition service system and achieve optimization of booth layout and visitor route planning. This work combines the Deep Belief Network (DBN) model with reinforcement learning (RL) algorithms and Random Forest (RF) algorithms to design and construct a digital exhibition service system. This work utilizes publicly available exhibition promotion and display channels, selecting four common types of exhibitions such as industry exhibitions. Each type chooses 10 different time periods and content formats of exhibition data, which are scattered and arranged into five exhibition datasets. The work introduces the RF algorithm as an auxiliary classifier, extracts the features through the DBN model, and uses quantitative indicators to evaluate the robustness of the model and the accuracy and personalization of the recommended results. Meanwhile, the learned features and patterns are input into the RL algorithm to verify the system's decision optimization effect. Results demonstrate that 1) Under perturbed data conditions, the system's accuracy average differs by only 0.4% from the original data conditions;the average F1 score differs by 0.003;and the average recall rate differs by only 0.1%. This indicates that the system exhibits good robustness when facing perturbed environments. 2) Cross-validation results show that the system maintains stable classification efficiency across different folds, with accuracy ranging from 82% to 89%. The average time consumption for each fold does not exceed 10ms, indicating that the system can efficiently classify different types of exhibition data. 3) Variance analysis results show that the p-values corresponding to five indicators-recommendation accuracy, recommendation coverage rate, personalized recommendation effect score, recommendation click-through rate, and user satisfaction-are 0.036, 0.027, 0.037, 0.046, and 0.039, respectively. They are all less than 0.05,

关键词： Biological system modeling Accuracy Data models Classification algorithms Industries Radio frequency Random forests Digital systems Decision making Optimization methods Deep belief network model reinforcement learning algorithm random forest algorithm digital exhibition service system decision optimization

来源：评论

学校读者我要写书评

暂无评论

Robust and efficient task scheduling for robotics applications with reinforcement learning

引用

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE 2024年第PartA期127卷

作者： Tejer, Mateusz Szczepanski, Rafal Tarczewski, Tomasz Nicolaus Copernicus Univ Inst Engn & Technol Grudziadzka 5 PL-87100 Torun Poland

Effective task scheduling can significantly impact on performance, productivity, and profitability in many real-world settings, such as production lines, logistics, and transportation systems. Traditional approaches to task scheduling rely on heuristics or simple rule-based methods. However, with the emergence of machine learning and artificial intelligence, there is growing interest in using these methods to optimize task scheduling. In particular, reinforcement learning is a promising task scheduling approach, because it can learn from experience and adapt to changing conditions. One step often missed or neglected is choosing optimal algorithm parameters and different ways the environment could be implemented. The study analyzes the performance possibilities of task scheduling using reinforcement learning. The deep analysis allows to select highly efficient environment models and Q-learning parameters. Moreover, automatic selection based on optimization algorithms has been proposed. Regardless of the selected optimal parameters, the resilience to environmental changes seems poor. The deducted analysis motivated the Authors to develop a novel Hybrid Q-learning approach. It allows to provide superior efficiency regardless of the environmental parameters.

关键词： Task scheduling Robotic arm Sorting problem reinforcement learning algorithm Performance analysis Predictive strategies Task scheduling Robotic arm Sorting problem reinforcement learning algorithm Performance analysis Predictive strategies

来源：评论

学校读者我要写书评

暂无评论

Automatic voltage control considering demand response: Approximatively completed observed Markov decision process-based reinforcement learning scheme

引用

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS 2024年 161卷

作者： Gu, Yaru Huang, Xueliang Southeast Univ Sch Elect Engn Nanjing Peoples R China

To fully utilize the voltage regulation capacity of flexible load and distributed generations (DGs), we propose a novel Approximatively Completed Observed Markov Decision Process-based (ACOMDP-based) reinforcement learning (RL) (namely, ACMRL) scheme for a multi-objective Automatic Voltage Control (AVC) problem considering Differential Increment Incentive Mechanism (DIIM)-based Incentive-Based Demand Response (IBDR). Firstly, we propose a DIIM to motivate high-flexibility consumers to achieve maximum potential in realtime voltage control while ensuring the best economy. Secondly, we characterize the multi-objective AVC problem as an ACOMDP model, transformed from the Partially Observable Markov Decision Process (POMDP) model, by introducing a novel hidden system state vector that incorporates the belief state, and the high confidence probability vector. The belief state and the high-confidence probability vector describe the probability distribution extracted from the historical observed state, portraying the precise state and the uncertainty existing in the state update process. Then, the ACOMDP block is inputted into the RL block, which adopts a modified underlying network architecture with the Asynchronous Advantage Actor-Critic (MA3C) algorithm embedded with the Shared Modular Policies(SMP) module. The MA3C-based RL block, characterized by enhanced communication efficiency, enables expedited generation of optimal decision-making actions even in the face of substantial uncertainty. Case studies are conducted in a practical district in Suzhou, China, and simulation results validate the superior performance of the proposed methodology.

关键词： Automatic voltage control Partially observable system Uncertainty Differential increment incentive mechanism reinforcement learning algorithm

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 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

6Search: A reinforcement learning-based traceroute approach for efficient IPv6 topology discovery

引用

COMPUTER NETWORKS 2023年第1期235卷

作者： Liu, Ning Jia, Chunbo Hou, Bingnan Hou, Changsheng Chen, Yingwen Cai, Zhiping Natl Univ Def Technol Coll Comp Changsha 410073 Hunan Peoples R China

Topology discovery can infer the interconnection relationship between network entities. A complete network topology is of great significance for network security analysis, application research, etc. However, due to the huge address space and uneven distribution of active addresses in the IPv6 Internet, it is infeasible to use brute-force traceroute to discover the entire topology. To address this problem, we propose 6Search, a target generation method based on reinforcement learning algorithm for IPv6 topology discovery. 6Search first obtains the routeable BGP prefixes and then carries out traceroute in each (/32) prefix. The number of probes allocated is dynamically adjusted based on the results of previous scans. Using the reinforcement learning algorithm, 6Search allocates more probes to prefixes with more address discovery in each scan iteration. Real world experiments demonstrate that 6Search has better performance in terms of discovery efficiency, which is 24.1%-139.8% improvement over the existing methods.

关键词： IPv6 Topology discovery reinforcement learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Optimized adaptive event-triggered tracking control for multi-agent systems with full-state constraints

引用

INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR CONTROL 2022年第18期32卷 10101-10124页

作者： Yang, Xiaoyu Pan, Yingnan Sun, Jize Tan, Lihua Bohai Univ Coll Control Sci & Engn Jinzhou 121013 Liaoning Peoples R China Shenyang Aircraft Design & Res Inst Shenyang Liaoning Peoples R China Southwest Univ Coll Elect & Informat Engn Chongqing Peoples R China

In this article, the event-triggered optimized adaptive tracking control problem is investigated for a class of multi-agent systems subject to full-state constraints. To address the full-state constraints problem, a nonlinear mapping technique is applied, which can release the feasibility conditions for virtual controllers. Based on the mean-value theorem, the nonaffine nonlinear terms generated by transformed system are separated, which overcomes the obstacle of solving optimized solution. The neural network based reinforcement learning (RL) algorithm with the identifier-critic-actor architecture is introduced to obtain the optimal solution of systems with unknown dynamics. It is worth noting that the RL algorithm in this article is simplified, which can reduce the computational burden. To reducing the communication burden, an event-triggered mechanism with time-varying threshold related to the optimized control signal is developed. By applying the Lyapunov stability method, it is proved that the desired optimized tracking performance and the stability of the closed-loop systems can be guaranteed. Finally, a simulation example demonstrates that the proposed control strategy is effective.

关键词： event-triggered control full-state constraints multi-agent systems reinforcement learning algorithm

来源：评论

学校读者我要写书评

暂无评论

Speed harmonisation and merge control using connected automated vehicles on a highway lane closure: a reinforcement learning approach

引用

IET INTELLIGENT TRANSPORT SYSTEMS 2020年第8期14卷 947-958页

作者： Ko, Byungjin Ryu, Seunghan Park, Byungkyu Brian Son, Sang Hyuk Daegu Gyeongbuk Inst Sci & Technol DGIST Dept Informat & Commun Engn 333 Techno Jungang Daero Daegu South Korea Univ Virginia Link Lab POB 400747 Charlottesville VA 22903 USA Univ Virginia Dept Engn Syst & Environm POB 400747 Charlottesville VA 22903 USA

A lane closure bottleneck usually leads to traffic congestion and a waste of fuel consumption on highways. In mixed traffic that consists of human-driven vehicles and connected automated vehicles (CAVs), the CAVs can be used for traffic control to improve the traffic flow. The authors propose speed harmonisation and merge control, taking advantage of CAVs to alleviate traffic congestion at a highway bottleneck area. To this end, they apply a reinforcement learning algorithm called deep Q network to train behaviours of CAVs. By training the merge control Q-network, CAVs learn a merge mechanism to improve the mixed traffic flow at the bottleneck area. Similarly, speed harmonisation Q-network learns speed harmonisation to reduce fuel consumption and alleviate traffic congestion by controlling the speed of following vehicles. After training two Q-networks of the merge mechanism and speed harmonisation, they evaluate the trained Q-networks under various conditions in terms of vehicle arrival rates and CAV market penetration rates. The simulation results indicate that the proposed approach improves the mixed traffic flow by increasing the throughput up to 30% and reducing the fuel consumption up to 20%, when compared to the late merge control without speed harmonisation.

关键词： traffic engineering computing road vehicles road traffic control road traffic traffic control learning (artificial intelligence) speed harmonisation connected automated vehicles highway lane closure reinforcement learning approach lane closure bottleneck traffic congestion fuel consumption human-driven vehicles traffic control highway bottleneck area reinforcement learning algorithm Q-network mixed traffic flow trained Q-networks vehicle arrival rates CAV market penetration rates

来源：评论

学校读者我要写书评

暂无评论

Dynamic spectrum access-based augmenting coverage in narrow band Internet of Things

引用

INTERNATIONAL JOURNAL OF COMMUNICATION SYSTEMS 2024年第1期37卷

作者： Gnanaselvam, R. Vasanthi, M. S. SRM Inst Sci & Technol Dept Elect & Commun Engn Kattankulathur Tamil Nadu India

The Third Generation Partnership Project launched the Narrowband Internet of Things (NB-IoT) as part of 5G technology under Release-13. These networks are authorized Low Power Wide Area Networks. It uses less power and sends message over long distances. The NB-IoT is can be used widely in industries, environment, and home for automation. While comparing NB-IoT to regular Long Term Evaluation Machine Type Communication, the repetition in uplink is 128. The extreme coverage by 164 dB in terms of Maximum Coupling Loss. The NB-IoT is employed in deep coverage for users. If we use many repetitions in NB-IoT technology, the results are shorter battery life and higher maintenance cost. Hence, we suggest abandoning the random spectrum access strategy in favor of a novel technique termed Dynamic Spectrum Access, which uses reinforcement learning algorithm to eliminate the shortcomings. This method is used to decrease the number of re-transmissions so that NB-IoT system power goes to less and augments coverage. The Third Generation Partnership Project launched the Narrowband Internet of Things (NB-IoT) as part of 5G technology under Release-13. These networks are authorized Low Power Wide Area Networks. It uses less power and sends message over long distances. The NB-IoT can be used widely in industries, environment, and home for ***

关键词： coverage improvement MAB model NB-IoT Proposed Dynamic Spectrum Access procedure reinforcement learning algorithm

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：