检索结果-内蒙古大学图书馆

2021 International symposium of Asian control Association on Intelligent Robotics and Industrial automation, IRIA 2021

作者： Ali, Zunaib Putrus, Ghanim Marzband, Mousa Tookanlou, Mahsa Bagheri Saleem, Komal Ray, Pravat Kumar Subudhi, Bidyadhar University of Northumbria Department of Mathematics Physics and Electrical Engineering Newcastle United Kingdom Queen Mary University of London School of Electrical and Computer Engineering London United Kingdom Department of Electrical Engineering Rourkela India Department of Electrical Engineering Goa India

ISBN: (纸本)9781665433235

Meteorological conditions such as air density, temperature, solar radiation etc. strongly affect the power generation from solar, and thus, the prediction and estimation process should consider weather conditions as critical inputs. The nature of weather forecast is highly unpredictable, so many applications use meteorological data from in-place on-site sensors to add to the forecast and some use complex networks with complicated mapping. The in-situ sensor approach and dense mapping methods, however, present several drawbacks. First, the use of sensors give rise to extra operational, installation and maintenance cost. Second, it requires significant amount of time to capture and accumulate data for various occasions and scenarios, and in addition, sensor itself can be the cause of error measurements. The complex methods are computational inefficient and may present suboptimal convergence. This paper presents a sensorless solar output power forecasting based on historical weather (publicly available from met office) and PV data. The algorithm uses simple to implement neural networks with few neurons and hidden layers for its training and allows for day a head forecast. The proposed methodology presents a guideline on how to select the relevant data from weather and how it affects the accuracy and training time of neural network. © 2021 ieee.

关键词： Solar energy

来源：评论

学校读者我要写书评

暂无评论

Artificial Identification, Blockchain, Cyberphysical Social Systems, Digital Twins, and Parallel intelligence Opportunities and Synergies Between the ieee Council on Radio-Frequency Identification and Systems, Man, and Cybernetics Society

引用

ieee SYSTEMS MAN AND CYBERNETICS MAGAZINE 2021年第2期7卷 61-66页

作者： Wang, Fei-Yue Rudas, Imre J. Wu, Dongrui Wang, Xiao Yuan, Yong Zhang, Jun Jason Li, Yidong Bennett, Gisele Bassiri-Gharb, Nazanin Univ Arizona Tucson AZ 85721 USA Univ Arizona Robot & Automat Lab Tucson AZ 85721 USA Univ Arizona Program Adv Res Complex Syst Tucson AZ 85721 USA Chinese Acad Sci Intelligent Control & Syst Engn Ctr Inst Automat Beijing Peoples R China Chinese Acad Sci Key Lab Complex Syst & Intelligence Sci Beijing Peoples R China State Key Lab Management & Control Complex Syst Beijing 100190 Peoples R China CAAs Supervis Council Beijing Peoples R China IEEE Council Radiofrequency Identificat Beijing Peoples R China IEEE Syst Man & Cybernet Soc Beijing Peoples R China Obuda Univ H-1034 Budapest Hungary Huazhong Univ Sci & Technol Artificial Intelligence & Automat Wuhan 430074 Hubei Peoples R China Chinese Acad Sci State Key Lab Management & Control Complex Syst Inst Automat Beijing 100190 Peoples R China Qingdao Acad Intelligent Ind Qingdao Qingdao 266109 Peoples R China Renmin Univ China Sch Math Beijing 100872 Peoples R China Minist Educ Engn Res Ctr Finance Computat & Digital Engn Beijing Peoples R China Wuhan Univ Sch Elect Engn & Automat Wuhan 430072 Peoples R China Beijing Jiaotong Univ Sch Comp & Informat Technol Beijing 100044 Peoples R China MEPSS LLC Indian Harbor Beach FL 32937 USA Georgia Tech Atlanta GA 30332 USA Georgia Inst Technol GW Woodruff Sch Mech Engn Atlanta GA 30332 USA

This article describes several opportunities and synergies between the ieee Council on Radio-Frequency Identification (CRFID) and ieee Systems, Man, and Cybernetics Society (SMCS) to initiate a roadmap study and working plan for a new model of support and collaboration among ieee Societies and Councils in the future. We hope this will stimulate more communication and discussion for deep and effective coordination and collaboration among ieee Councils and Societies.

关键词： Digital twin Collaboration Blockchain Cybernetics Radiofrequency identification

来源：评论

学校读者我要写书评

暂无评论

An Improved YOLOv5 Model for Detecting Citrus Shoots in Complex Orchard Environments

An Improved YOLOv5 Model for Detecting Citrus Shoots in Comp...

引用

Robotics, automation and Artificial intelligence (RAAI), ieee International Conference on

作者： Feng Xiao Haibin Wang Tianqi Huang Yueqin Xu College of Mechanical and Electrical Engineering Northeast Forestry University Harbin China

ISBN: (数字)9798350307566

ISBN: (纸本)9798350307573

The development of citrus shoots is of great significance for the growth, flowering, fruiting, and pest and disease control of citrus trees. In order to achieve accurate recognition of citrus shoots in complex orchard environments, an improved YOLOv5 model, named YOLOv5-RepVGG-SE, was proposed. Firstly, in order to enhance the feature extraction capability, the Rep VGG modules are used to replace the Conv modules in the backbone networks of the YOLOv5 model. Second, the Squeeze-Excitation (SE) attention modules are added to the neck networks of the YOLOv5 model to improve the feature fusion capability. A dataset of citrus shoots was established to evaluate the effectiveness of the improved YOLOv5 model. The experimental results show that the improved YOLOv5 model achieves a precision ( $P$ ) of 81.1 %, a recall ( $R$ ) of 76.8%, and a mean average precision (mAP) of 71.0% at a threshold of 0.5. These values are 0.6%, 9.20/0, and 9.4% higher, respectively, than those of the YOLOv5 model. This research can serve as a reference for the development of smart devices in citrus orchards.

关键词： YOLO Accuracy Phenotypes computational modeling Vegetation Feature extraction Neck

来源：评论

学校读者我要写书评

暂无评论

Collision-free vector field guidance and MPC for a fixed-wing UAV

Collision-free vector field guidance and MPC for a fixed-win...

引用

ieee International Conference on Robotics and automation (ICRA)

作者： Pereira, Leonardo A. A. Nunes, Arthur H. D. Rezende, Adriano M. C. Goncalves, Vinicius M. Raffo, Guilherme, V Pimenta, Luciano C. A. Univ Fed Minas Gerais Grad Program Elect Engn Av Antonio Carlos 6627 BR-31270901 Belo Horizonte MG Brazil

ISBN: (纸本)9781728190778

The present work focuses on the development of an efficient path controller to guide a fixed-wing UAV (Unmanned Aerial Vehicle) to follow a closed curve and avoid unknown dynamic obstacles. Our strategy is composed of two layers: a top level layer responsible for guidance and a lower level layer responsible for tracking the references given by the top level. To solve the guidance problem, we propose a vector field strategy that switches between two forms: a vector field to converge and circulate the target curve and a vector field to avoid obstacles by circulating the closest one. To make the fixed-wing UAV follow the velocity provided by the guidance vector field we consider a Model Predictive control scheme. The feedback linearization allows efficient computation of control commands as a linear MPC controller can be employed. Our results are validated in simulations that take into account the 6DOF (Degrees of Freedom) model with constraints of the aircraft, wind disturbance and uncertainties on the measurements.

关键词： Uncertainty Target tracking Shape computational modeling Atmospheric modeling Measurement uncertainty Unmanned aerial vehicles

来源：评论

学校读者我要写书评

暂无评论

Neural-Physical Fusion Computation for Container Terminal Handling Systems by computational Logistics and Deep Learning

Neural-Physical Fusion Computation for Container Terminal Ha...

引用

ieee International Conference on Systems, Man, and Cybernetics (SMC)

作者： Li, Bin He, Yuqing Li, Wenfeng Fujian Univ Technol Sch Mech & Automot Engn Fuzhou 350118 Fujian Peoples R China Fujian Univ Technol Sch Transportat Fuzhou 350118 Fujian Peoples R China Wuhan Univ Technol Sch Logist Engn Wuhan 430063 Hubei Peoples R China

ISBN: (纸本)9781665442077

The emerging information technology and the rising computational thinking make it possible to establish new theoretical methods and engineering practices for planning, scheduling, control and decision-making of complex logistics systems. Consequently, the complex logistics systems oriented neural-physical fusion computation (CLSO-NPFC) is proposed initially to discuss and explore the decision issues of complex logistics systems at all the strategic, tactical and operational levels. By CLSO-NPFC, the container terminal oriented logistics generalized computing mechanization, automation and intelligence are constructed uniformly and tentatively, and the typical deep learning model neural computing architecture (DLM-NCA) in CLSO-NPFC is designed for the prediction of calling liner handling volume that is measured by container units rather than twenty-feet equivalent unit. A typical regional container terminal along the coast of China is selected to implement, execute and evaluate the DLM-NCA, and the DLM-NCA shows the agile, efficient and robust performance for forecasting with the low and stable computation consumption. It demonstrates the feasibility, credibility and practicality of the abstract principles, design paradigms and computing architecture in CLSO-NPFC preliminarily.

关键词： Containers

来源：评论

学校读者我要写书评

暂无评论

DHC-R: Evaluating "Distributed Heuristic Communication" and Improving Robustness for Learnable Decentralized PO-MAPF 1

引用

Proceedings of the 8th International Conference on Interactive Collaborative Robotics, ICR 2023

作者： Savinov, Vladislav Yakovlev, Konstantin St Petersburg State University 7-9 Universitetskaya Emb. St. Petersburg199034 Russia Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences 44/2 Vavilova street Moscow119333 Russia

ISBN: (数字)9783031431111

ISBN: (纸本)9783031431104

Multi-agent pathfinding (MAPF) is a problem of coordinating the movements of multiple agents operating a shared environment that has numerous industrial and research applications. In many practical cases the agents (robots) have limited visibility of the environment and must rely on local observations to make decisions. This scenario, known as partially observable MAPF (PO-MAPF), can be solved through decentralized approaches. In recent years, several learnable algorithms have been proposed for solving PO-MAPF. However, their performance is oftentimes not validated out-of-distribution (OOD), and the code is often not properly open-sourced. In this study, we conduct a comprehensive empirical evaluation of one of the state-of-the-art decentralized PO-MAPF algorithms, Distributed Heuristic Communication (DHC), Ma, Z., Luo, Y., Ma, H.: Distributed heuristic multi-agent path finding with communication. In: 2021 International Conference on Robotics and automation (ICRA), pp. 8699–8705. ieee, Xi’an, China (2021), which incorporates communication between agents. Our experiments reveal that the performance of DHC deteriorates when agents encounter complete packet loss during communication. To address this issue, we propose a novel algorithm called DHC-R that employs a similar architecture to the original DHC but introduces randomness into the graph neural network-based communication block, preventing the passage of some data packets during training. Empirical evaluation confirms that DHC-R outperforms DHC in scenarios with packet loss. Open-sourced model weights and the codebase are provided: https://***/acforvs/dhc-robust-mapf. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Scaling, control and Generalization in Reinforcement Learning Level Generators

Scaling, Control and Generalization in Reinforcement Learnin...

引用

ieee symposium on computational intelligence and Games, CIG

作者： Sam Earle Zehua Jiang Julian Togelius Game Innovation Lab New York University NY Brooklyn

ISBN: (数字)9798350350678

ISBN: (纸本)9798350350685

Procedural Content Generation via Reinforcement Learning (PCGRL) has been introduced as a means by which controllable designer agents can be trained based only on a set of computable metrics acting as a proxy for the level’s quality and key characteristics. While PCGRL offers a unique set of affordances for game designers, it is constrained by the compute-intensive process of training RL agents, and has so far been limited to generating relatively small levels. To address this issue of scale, we implement several PCGRL environments in Jax so that all aspects of learning and simulation happen in parallel on the GPU, resulting in faster environment simulation; removing the CPU-GPU transfer of information bottleneck during RL training; and ultimately resulting in significantly improved training speed. We replicate several key results from prior works in this new framework, letting models train for much longer than previously studied, and evaluating their behavior after 1 billion timesteps. Aiming for greater control for human designers, we introduce randomized level sizes and frozen “pinpoints” of pivotal game tiles as further ways of countering overfitting. To test the generalization ability of learned generators, we evaluate models on large, out-of-distribution map sizes, and find that partial observation sizes learn more robust design strategies.

关键词： Training Measurement Procedural generation Limiting Shape Layout Graphics processing units

来源：评论

学校读者我要写书评

暂无评论

Online Dynamic Trajectory Optimization and control for a Quadruped Robot

Online Dynamic Trajectory Optimization and Control for a Qua...

引用

ieee International Conference on Robotics and automation (ICRA)

作者： Cebe, Oguzhan Tiseo, Carlo Xin, Guiyang Lin, Hsiu-chin Smith, Joshua Mistry, Michael Univ Edinburgh Inst Percept Act & Behav Sch Informat 10 Crichton St Edinburgh EH8 9AB Midlothian Scotland McGill Univ Dept Elect & Comp Engn Sch Comp Sci Montreal PQ Canada

ISBN: (纸本)9781728190778

Legged robot locomotion requires the planning of stable reference trajectories, especially while traversing uneven terrain. The proposed trajectory optimization framework is capable of generating dynamically stable base and footstep trajectories for multiple steps. The locomotion task can be defined with contact locations, base motion or both, making the algorithm suitable for multiple scenarios (e.g., presence of moving obstacles). The planner uses a simplified momentum-based task space model for the robot dynamics, allowing computation times that are fast enough for online replanning. This fast planning capability also enables the quadruped to accommodate for drift and environmental changes. The algorithm is tested on simulation and a real robot across multiple scenarios, which includes uneven terrain, stairs and moving obstacles. The results show that the planner is capable of generating stable trajectories in the real robot even when a box of 15 cm height is placed in front of its path at the last moment.

关键词： Legged locomotion computational modeling Heuristic algorithms Conferences Dynamics Stairs Planning

来源：评论

学校读者我要写书评

暂无评论

Cross-Domain Machine Learning Approaches using Hyperspectral Imaging for Plastics litter detection

Cross-Domain Machine Learning Approaches using Hyperspectral...

引用

Geoscience and Remote Sensing symposium (M2GARSS), Mediterranean and Middle-East

作者： Soufyane Bouchelaghem Imad Eddine Tibermacine Marco Balsi Monica Moroni Christian Napoli Dipartimento di Ingegneria Dell’Informazione Elettronica e Telecomunicazioni (DIET) Sapienza University of Rome Rome Italy Department of Computer Automation and Management Engineering (DIAG) Sapienza University of Rome Rome Italy Dipartimento di Ingegneria Civile Edile e Ambientale Sapienza University of Rome Rome Italy Institute for Systems Analysis and Computer Science Italian National Research Council Rome Italy Department of Computational Intelligence Czestochowa University of Technology Czestochowa Poland

ISBN: (数字)9798350358582

ISBN: (纸本)9798350358599

Plastics detection and polymer separation are important tasks for environmental monitoring and waste recycling. The authors previously used hyperspectral data in the lower bands of the SWIR range from 900nm to 1700 nm to perform plastics classification, by applying supervised machine learning methods to data gathered by a custom-assembled stand-alone push-broom system in indoor and outdoor conditions. In this work, we explored the application of two unsupervised learning algorithms: K-Means clustering and Gaussian Mixture Models (GMM), in order to avoid the use of a manually-labeled test set, and to obtain more reliable and flexible application in different environmental conditions. The results suggest that GMM performs well indoors but poorly outside, with GMM exhibiting subtle segmentation and K-Means displaying broader, more generalist categorization. The cross-domain adaptation is difficult and models that balance granularity and generalization are needed to detect plastic in varied settings. This effort advances garbage management and recycling and establishes hyperspectral imaging in real-world environments.

关键词： Image segmentation Adaptation models Machine learning algorithms Clustering algorithms Machine learning Recycling Plastics

来源：评论

学校读者我要写书评

暂无评论

Pattern Recognition with Antiferromagnet-heavy Metal Hybrid Structure

Pattern Recognition with Antiferromagnet-heavy Metal Hybrid ...

引用

Photonics and Electromagnetics Research symposium (PIERS)

作者： Mitrofanova, A. Yu Safin, A. R. Egorov, D. P. Kravchenko, O., V Bazenkov, N., I Bauman Moscow State Tech Univ 2 Baumanskaya St Moscow 105005 Russia Russian Acad Sci Kotelnikov Inst Radioengn & Elect Mokhovaya 11-7 Moscow 125009 Russia Moscow Power Engn Inst 14 Krasnokazarmennaya St Moscow 111250 Russia RAS Fed Res Ctr Comp Sci & Control Vavilova St 40 Moscow 119333 Russia Russian Acad Sci VA Trapeznikov Inst Control Sci 65 Profsoyuznaya St Moscow 117997 Russia

ISBN: (纸本)9781728172477

Nowadays, there is a great interest in artificial intelligence. One of its popular branches is deep learning, based on the neural networks consisting of artificial neurons, which are adders with a nonlinear activation function. This approach requires large amounts of training data. For example, computer vision contains different tasks of this nature, where one of the well-known problems is recognizing images of handwritten numbers. Various approaches to training convolutional neural networks on the MNIST database show less than 0.5% error. Another area of artificial intelligence is neuromorphic computing, whose goal is to build biologically plausible models of neurons in the human brain. The motivation for the approach of neuromorphic computation is that accurate modeling of biological neurons will allow solving many problems, in particular, control problems, with the same accuracy the human does. Moreover, it opens up the possibility of creating electrical circuits that mimic the behavior of biological neurons, which will reduce energy consumption. In particular, the physical realization of neurons can be coupled oscillators that are ubiquitous in nature. The phase or frequency of the oscillators contains information that allows development of such computational schemes. In this paper, a hybrid antiferromagnetic-heavy metal structure is used to solve the problem of image recognition. During the model training the coupling coefficients of the oscillators are calculated so that pairs of oscillators corresponding to different pixel values oscillate asynchronously, and the oscillators reacting to the same value set the oscillation mode in phase. Pattern recognition error is estimated depending on the number of neurons, patterns, and signal-to-noise ratio (SNR).

关键词： Neurons

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：