检索结果-内蒙古大学图书馆

DNN Task Assignment in UAV Networks: A Generative AI Enhanced Multi-Agent Reinforcement Learning Approach

学校读者我要写书评

暂无评论

arXiv

arXiv 2024年

作者： Tang, Xin Chen, Qian Weng, Wenjie Liao, Binhan Wang, Jiacheng Cao, Xianbin Li, Xiaohuan Guangxi University Key Laboratory of Intelligent Networking and Scenario System School of Information and Communication Guilin University of Electronic Technology Guilin541004 China The College of Computing and Data Science Nanyang Technological University 639798 Singapore School of Architecture and Transportation Engineering Guilin University of Electronic Technology Guilin541004 China National Engineering Laboratory for Comprehensive Transportation Big Data Application Technology Guangxi Nanning530001 China College of Computing and Data Science Nanyang Technological University 639798 Singapore School of Electronic and Information Engineering The Key Laboratory of Advanced Technology of Near Space Information System Ministry of Industry and Information Technology of China Beihang University Beijing100191 China

Unmanned Aerial Vehicles (UAVs) possess high mobility and flexible deployment capabilities, prompting the development of UAVs for various application scenarios within the Internet of Things (IoT). The unique capabilities of UAVs give rise to increasingly critical and complex tasks in uncertain and potentially harsh environments. The substantial amount of data generated from these applications necessitates processing and analysis through deep neural networks (DNNs). However, UAVs encounter challenges due to their limited computing resources when managing DNN models. This paper presents a joint approach that combines multiple-agent reinforcement learning (MARL) and generative diffusion models (GDM) for assigning DNN tasks to a UAV swarm, aimed at reducing latency from task capture to result output. To address these challenges, we first consider the task size of the target area to be inspected and the shortest flying path as optimization constraints, employing a greedy algorithm to resolve the subproblem with a focus on minimizing the UAV’s flying path and the overall system cost. In the second stage, we introduce a novel DNN task assignment algorithm, termed GDM-MADDPG, which utilizes the reverse denoising process of GDM to replace the actor network in multi-agent deep deterministic policy gradient (MADDPG). This approach generates specific DNN task assignment actions based on agents’ observations in a dynamic environment. Simulation results indicate that our algorithm performs favorably compared to benchmarks in terms of path planning, Age of Information (AoI), energy consumption, and task load balancing. Copyright © 2024, The Authors. All rights reserved.

关键词： Motion planning

中国双碳目标下流程制造业生命周期工程评价方法及应用（英文）

学校读者我要写书评

暂无评论

engineering 2024年第9期 87-99页

作者：李明阳高峰聂祚仁孙博学刘宇龚先政 National Engineering Laboratory for Industrial Big-data Application Technology College of Material Science and Engineering Beijing University of Technology

The industrial sector is the primary source of carbon emissions in China. In pursuit of meeting its carbon reduction targets, China aims to promote resource consumption sustainability, reduce energy consumption, and achieve carbon neutrality within its processing industries. An effective strategy to promote energy savings and carbon reduction throughout the life cycle of materials is by applying life cycle engineering technology. This strategy aims to attain an optimal solution for material performance, resource consumption, and environmental impact. In this study, five types of technologies were considered: raw material replacement, process reengineering, fuel replacement, energy recycling and reutilization, and material recycling and reutilization. The meaning, methodology, and development status of life cycle engineering technology abroad and domestically are discussed in detail. A multidimensional analysis of ecological design was conducted from the perspectives of resource and energy consumption, carbon emissions, product performance, and recycling of secondary resources in a manufacturing process. This coupled with an integrated method to analyze carbon emissions in the entire life cycle of a material process industry was applied to the nonferrous industry, as an example. The results provide effective ideas and solutions for achieving low or zero carbon emission production in the Chinese industry as recycled aluminum and primary aluminum based on advanced technologies had reduced resource consumption and emissions as compared to primary aluminum production.

关键词：

A Modelling Approach for Coordinated Car Flow Assignment between Technical Stations Based on the Optimization of the Departure Mode Strategy

学校读者我要写书评

暂无评论

SSRN 2022年

作者： Xue, Feng Cheng, Daibing Yang, Zongqin Ma, Xiaochen Chen, Chongshuang School of Transportation and Logistics Southwest Jiaotong University Chengdu611756 China National Engineering Laboratory of Integrated Transportation Big Data Application Technology Southwest Jiaotong University Chengdu611756 China National United Engineering Laboratory of Integrated and Intelligent Transportation Southwest Jiaotong University Chengdu611756 China Transportation and Urban Planning Group Eindhoven University of Technology Eindhoven5612AZ Netherlands School of Mathematics Southwest Jiaotong University Chengdu611756 China

This study analysed the characteristics and departure conditions of different accumulation modes and design a freight train cooperative operation scheme between technical stations based on a hybrid accumulation mode. A coordinated freight train assignment model between technical stations is proposed with the objective function of minimizing the dwell time of cars at two technical stations. Considering the characteristics of the car flow connection problem, the standard grey wolf optimization algorithm coding method and individual location update formula are improved, and are used for optimization. The classification sequence and assignment plan of freight trains are obtained. The results of a case study show that the total dwell time of the vehicles at two technical stations has been reduced by 35.82 h, and the number of punctual outbound trains in the phase plan has been increased by two. The cooperative flow assignment improves the efficiency of the car flow organization compared with the single station flow assignment. © 2022, The Authors. All rights reserved.

关键词： Optimization

Navigation Line Extraction Algorithm for Corn Spraying Robot Based on Improved Yolov8s Network

学校读者我要写书评

暂无评论

SSRN 2023年

作者： Diao, Zhihua Guo, Peiliang Zhang, Baohua Zhang, Dongyan Yan, Jiaonan He, Zhendong Zhao, Suna Zhao, Chunjiang Zhang, Jingcheng School of Electrical Information Engineering Zhengzhou University of Light Industry Zhengzhou450002 China Nanjing211800 China National Engineering Research Center for Agro-Ecological Big Data Analysis & Application Anhui University Hefei230601 China Information Technology Research Center Beijing Academy of Agriculture and Forestry Sciences Beijing100097 China School of Automation Hangzhou Dianzi University Hangzhou310000 China

Aiming at the shortcomings of the existing navigation line extraction algorithm for corn spraying robot in complex farmland environment, such as poor extraction effect and poor adaptability, this paper proposed a navigation line extraction algorithm for corn spraying robot based on improved YOLOv8s network. This algorithm proposes to use corn plant core as the identification target to locate corn plants. Firstly, a new spatial pyramid structure ASPPF is proposed in this paper, and an improved YOLOv8s model ASPPF-YOLOv8s is proposed for more accurate detection of corn plant cores. Secondly, the center coordinates of the network detection box were used to locate the corn crop row feature points. Finally, the least squares method was used to fit the corn crop row lines. Experimental results showed that ASPPF-YOLOv8s network had a good effect on corn plant cores extraction under different growth periods and environmental pressures. MAP and F1 of ASPPF-YOLOv8s network increased from 86.4% and 86% of YOLOv7 network, 88.8% and 87% of YOLOv8s network, 88.6% and 89% of ASPP-YOLOv8s network to 90.2% and 91%. The average fitting time and average angle error of the centerline of crop row were reduced from 82.6ms and 0.97°for method of Zhang et al., 74.8ms and 0.75°for method of Hu et al., 67ms and 2.03°for method of Yang et al. to 45ms and 0.63°by using the center coordinates of corn plant core detection box combined with least square method. The accuracy rate increased from 91.4% for method of Zhang et al., 93.6% for method of Hu et al. and 87.35% for method of Yang et al. to 94.35%. It shows that the navigation line extraction algorithm proposed in this paper can meet the requirements of real-time and accuracy of visual navigation of corn spraying robot, and can be used to extract navigation line of corn spraying robot in complex farmland environment. © 2023, The Authors. All rights reserved.

关键词： Extraction

Improving Pedestrian Traffic by Sound-Induced Step Synchronization

学校读者我要写书评

暂无评论

SSRN 2022年

作者： Fu, Zhijian Schadschneider, Andreas Xiong, Xingwen Luo, Lin Jia, Qihao School of Transportation and Logistics Southwest Jiaotong University PRC Chengdu611756 China Institut für Theoretische Physik Universität zu Köln Köln50937 Germany National Engineering Laboratory for Integrated Transportation Big Data Application Technology Southwest Jiaotong University Chengdu611756 China National United Engineering Laboratory of Integrated and Intelligent Transportation Southwest Jiaotong University Chengdu611756 China

We develop a method at the operational level for pedestrian traffic improvement and seek a proper explanation for the discrepancy among the fundamental diagrams of pedestrian dynamics by experiments and theoretical analysis. Since sound is flexible and can strongly influence the walking behavior of pedestrians, in the experiment, participants were asked to walk under different types of sound. Results show that (1) regular sound increases pedestrian efficiency. Walking under a monotone with a rhythm slower than the normal walking pace, the pedestrian flow increases by 36.8% at most. Periodic tones have an even better performance than monotones, and the flow falls as pedestrians do not follow the sound. (2) Under regular tones, participants become less sensitive to the spatial distance, and step synchronization (i.e., marching in lock-step) becomes more likely to occur, diminishing stop-go movement. Then, the utilization of the available space between pedestrians in the longitudinal direction is significantly improved in the high-density regime. The theory is developed relating the macro-flow with the micro-step, and finally, possible engineering applications of this research are discussed. © 2022, The Authors. All rights reserved.

关键词： Synchronization

Accelerating and Securing Federated Learning via Intelligent Reflecting Surface

学校读者我要写书评

暂无评论

Accelerating and Securing Federated Learning via Intelligent...

International Conference on Communication technology (ICCT)

作者： Yingying Wu Bomin Mao School of Cybersecurity Northwestern Poly technical University Xi'an Shaanxi China National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology School of Cybersecurity Northwestern Poly technical University Xi'an China Yangtze River Delta Research Institute of Northwestern Poly technical University Taicang Jiangsu China

Federated Learning (FL) provides a privacy-protected way to train Machine Learning (ML) models, where multiple edge devices upload the model parameters to the central server without sharing their local data. However, the stochastic nature of wireless channels and frequent communications involved in FL incur high latency while model parameters can be intercepted by eavesdroppers, which leads to privacy leakage. To address the above issues, we exploit the Intelligent Reflecting Surface (IRS) to reconfigure the wireless signal propagation, enabling secure transmission and fast convergence. In this paper, We aim to maximize the minimum secrecy rate at devices via jointly designing transmission power at devices and IRS phase shift. A Deep Reinforcement Learning (DRL) based algorithm is adopted to derive the optimal solution. Transmission power and IRS phase shift are treated as the action elements. These elements are dynamically adjusted through interactions between the agent and the environment, guided by a pre-defined optimal reward. Both the transmission power and IRS phase shift are obtained as outputs from the DRL neural network. Numerical results validate the efficiency of our proposed algorithm and demonstrate that the deployment of IRS can improve the transmission rate and secure the training of FL.

关键词：

Synthesis of Flower-Like Bi/Bi2o3/Wo3 Tri-Composites and the Visible-Light-Driven Photocatalytic Property

学校读者我要写书评

暂无评论

SSRN 2022年

作者： Xi, Xiaoli Zhao, Linyan Liu, Yangsi Cheng, Jiali Nie, Zuoren Collaborative Innovation Center of Capital Resource-Recycling Material Technology Faculty of Materials and Manufacturing Beijing University of Technology Beijing100124 China Key Laboratory of Advanced Functional Materials Ministry of Education Beijing University of Technology Beijing100124 China National Engineering Laboratory for Industrial Big-data Application Technology Beijing University of Technology Beijing100124 China Beijing Guyue New Materials Research Institute Beijing University of Technology Beijing100124 China

A novel tri-composite Bi/Bi2O3/WO3 product was fabricated via a facile and green hydrothermal-solvothermal method, and the morphology structure was tailored by adjusting the dosage of EDTA. The final three-dimensional (3D) flower-like Bi/Bi2O3/WO3 is in structure with one-dimensional (1D) nanorods surrounded by two-dimensional (2D) nanoflakes and zero-dimensional (0D) Bi spheres dispersed around randomly. Moreover, the growth mechanism of the flower-like Bi/Bi2O3/WO3 was clarified and illustrated, where the functions of organic ethylene glycol and EDTA were considered, and a coordination and epitaxy growth process was proposed. The Bi/Bi2O3/WO3 composite performed highest visible-light photocatalytic activities to organic pollutant methylene blue (MB) of up to 95% degradation rate, and impressive photocatalytic efficiency to methyl orange (MO) in a tri-pollutants system, indicating the application potential in wastewater treatment. The superior photocatalytic performance could mainly contribute to the large proportion of flower-like structure and the synergy effect of the dual-path transfer process of charges, where the Z-scheme structure and localized surface plasmon resonance (LSPR) phenomenon is involved and discussed. © 2022, The Authors. All rights reserved.

关键词： Bismuth compounds

Preparation of Tungsten Molybdenum Oxide from Secondary Resources by a Facile Heat-Treatment Process and its Superior Selective Adsorption Properties

学校读者我要写书评

暂无评论

SSRN 2022年

作者： Liu, Yanan Liu, Yangsi Ma, Liwen Xi, Xiaoli Nie, Zuoren Collaborative Innovation Center of Capital Resource-Recycling Material Technology Beijing University of Technology Beijing100124 China Key Laboratory of Advanced Functional Materials Ministry of Education Faculty of Materials and Manufacturing Beijing University of Technology Beijing100124 China National Engineering Laboratory for Industrial Big-data Application Technology Beijing University of Technology Beijing100124 China Beijing Guyue New Materials Research Institute Beijing University of Technology Beijing100124 China

Tungsten molybdenum oxide was prepared by a facile heat-treatment process using ion exchange resins as the template. The morphology and crystallinity of the product can be controlled via varying the calcination temperature. The products show excellent selective adsorption properties for methylene blue (MB) and rhodamine B (RhB) model dyes with different molecular sizes and charges, especially for MB. The kinetic data show that the product accords with the pseudo-second-order kinetic model, and the adsorption isotherm can be described by Langmuir adsorption isotherm. When the concentration of MB is 100 mg/L, the adsorption capacity of the product for MB can be 833.43 mg/g, which is higher than that reported in many works of literature. Electrostatic adsorption was found to be the main adsorption mechanism. The heat-treatment method was successfully used to obtained tungsten molybdenum oxide from W-Mo alloy scrap. © 2022, The Authors. All rights reserved.

关键词： Tungsten compounds

Accurate spectral super-resolution from single RGB image using multi-scale CNN

学校读者我要写书评

暂无评论

arXiv

arXiv 2018年

作者： Yan, Yiqi Zhang, Lei Li, Jun Zhang, Yanning Wei, Wei School of Electronics and Information Northwestern Polytechnical University Xi'an China School of Computer Science Northwestern Polytechnical University Xi'an China National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology China Guangdong Provincial Key Laboratory of Urbanization and Geo-simulation School of Geography and Planning Sun Yat-Sen University Guangzhou China

Different from traditional hyperspectral super-resolution ap-proaches that focus on improving the spatial resolution, spectral super-resolution aims at producing a high-resolution hyperspectral image from the RGB observation with super-resolution in spectral domain. However, it is challenging to accurately reconstruct a high-dimensional continuous spectrum from three discrete intensity values at each pixel, since too much information is lost during the procedure where the latent hyper-spectral image is downsampled (e.g., with ×10 scaling factor) in spectral domain to produce an RGB observation. To address this problem, we present a multi-scale deep convolutional neural network (CNN) to ex-plicitly map the input RGB image into a hyperspectral image. Through symmetrically downsampling and upsampling the intermediate feature maps in a cascading paradigm, the local and non-local image informa-tion can be jointly encoded for spectral representation, ultimately im-proving the spectral reconstruction accuracy. Extensive experiments on a large hyperspectral dataset demonstrate the effectiveness of the pro-posed method. Copyright © 2018, The Authors. All rights reserved.

关键词： Optical resolving power