检索结果-内蒙古大学图书馆

GPT-4 enhanced multimodal grounding for autonomous driving:Leveraging cross-modal attention with large language models

引用

Communications in Transportation Research 2024年第1期4卷 5-23页

作者： Haicheng Liao Huanming Shen Zhenning Li Chengyue Wang Guofa Li Yiming Bie Chengzhong Xu State Key Laboratory of Internet of Things for Smart City and Department of Computer and Information Science University of MacaoMacao SAR999078China Department of Information and Software Engineering University of Electronic Science and Technology of ChinaChengdu610000China State Key Laboratory of Internet of Things for Smart City and Departments of Civil and Environmental Engineering and Computer and Information Science University of MacaoMacao SAR999078China State Key Laboratory of Internet of Things for Smart City and Departments of Civil and Environmental Engineering University of MacaoMacao SAR999078China College of Mechanical and Vehicle Engineering Chongqing UniversityChongqing400030China School of Transportation Jilin UniversityChangchun130000China

In the field of autonomous vehicles(AVs),accurately discerning commander intent and executing linguistic commands within a visual context presents a significant *** paper introduces a sophisticated encoder-decoder framework,developed to address visual grounding in *** Context-Aware Visual Grounding(CAVG)model is an advanced system that integrates five core encoders—Text,Emotion,Image,Context,and Cross-Modal—with a multimodal *** integration enables the CAVG model to adeptly capture contextual semantics and to learn human emotional features,augmented by state-of-the-art Large Language Models(LLMs)including *** architecture of CAVG is reinforced by the implementation of multi-head cross-modal attention mechanisms and a Region-Specific Dynamic(RSD)layer for attention *** architectural design enables the model to efficiently process and interpret a range of cross-modal inputs,yielding a comprehensive understanding of the correlation between verbal commands and corresponding visual *** evaluations on the Talk2Car dataset,a real-world benchmark,demonstrate that CAVG establishes new standards in prediction accuracy and operational ***,the model exhibits exceptional performance even with limited training data,ranging from 50%to 75%of the full *** feature highlights its effectiveness and potential for deployment in practical AV ***,CAVG has shown remarkable robustness and adaptability in challenging scenarios,including long-text command interpretation,low-light conditions,ambiguous command contexts,inclement weather conditions,and densely populated urban environments.

关键词： Autonomous driving Visual grounding Cross-modal attention Large language models Human-machine interaction

来源：评论

学校读者我要写书评

暂无评论

Review of Recent Trends in the Hybridisation of Preprocessing-Based and Parameter Optimisation-Based Hybrid Models to Forecast Univariate Streamflow

引用

computer Modeling in engineering & sciences 2024年第1期138卷 1-41页

作者： Baydaa Abdul Kareem Salah L.Zubaidi Nadhir Al-Ansari Yousif Raad Muhsen Department of Civil Engineering University of MaysanMaysan57000Iraq Department of Civil Engineering Wasit UniversityWasit52001Iraq College of Engineering University ofWarith Al-AnbiyaaKarbala56001Iraq Department of Civil Environmental and Natural Resources Engineering Lulea University of TechnologyLulea97187Sweden Department of Computer Science Faculty of Computer Science and Information TechnologyUniversiti Putra MalaysiaSerdang43400Malaysia

Forecasting river flow is crucial for optimal planning,management,and sustainability using freshwater *** machine learning(ML)approaches have been enhanced to improve streamflow *** techniques have been viewed as a viable method for enhancing the accuracy of univariate streamflow estimation when compared to standalone *** researchers have also emphasised using hybrid models to improve forecast ***,this paper conducts an updated literature review of applications of hybrid models in estimating streamflow over the last five years,summarising data preprocessing,univariate machine learning modelling strategy,advantages and disadvantages of standalone ML techniques,hybrid models,and performance *** study focuses on two types of hybrid models:parameter optimisation-based hybrid models(OBH)and hybridisation of parameter optimisation-based and preprocessing-based hybridmodels(HOPH).Overall,this research supports the idea thatmeta-heuristic approaches precisely improveML ***’s also one of the first efforts to comprehensively examine the efficiency of various meta-heuristic approaches(classified into four primary classes)hybridised with ML *** study revealed that previous research applied swarm,evolutionary,physics,and hybrid metaheuristics with 77%,61%,12%,and 12%,***,there is still room for improving OBH and HOPH models by examining different data pre-processing techniques and metaheuristic algorithms.

关键词： Univariate streamflow machine learning hybrid model data pre-processing performance metrics

来源：评论

学校读者我要写书评

暂无评论

High‑Order ADER Discontinuous Galerkin Schemes for a Symmetric Hyperbolic Model of Compressible Barotropic Two‑Fluid Flows

引用

Communications on Applied Mathematics and Computation 2024年第4期6卷 2119-2154页

作者： Laura Río‑Martín Michael Dumbser Department of Civil Environmental and Mechanical EngineeringUniversity of TrentoVia Mesiano 77Trento 38123Italy Department of Information Engineering and Computer Science University of Trentovia Sommarive 9PovoTrento 38123Italy

This paper presents a high-order discontinuous Galerkin(DG)finite-element method to solve the barotropic version of the conservative symmetric hyperbolic and thermodynamically compatible(SHTC)model of compressible two-phase flow,introduced by Romenski et ***[59,62],in multiple space *** the absence of algebraic source terms,the model is endowed with a curl constraint on the relative velocity *** this paper,the hyperbolicity of the system is studied for the first time in the multidimensional case,showing that the original model is only weakly hyperbolic in multiple space *** restore the strong hyperbolicity,two different methodologies are used:(i)the explicit symmetrization of the system,which can be achieved by adding terms that contain linear combinations of the curl involution,similar to the Godunov-Powell terms in the MHD equations;(ii)the use of the hyperbolic generalized Lagrangian multiplier(GLM)curl-cleaning approach *** PDE system is solved using a high-order ADER-DG method with a posteriori subcell finite-volume limiter to deal with shock waves and the steep gradients in the volume fraction commonly appearing in the solutions of this type of *** illustrate the performance of the method,several different test cases and benchmark problems have been run,showing the high order of the scheme and the good agreement when compared to reference solutions computed with other well-known methods.

关键词： Compressible two-fluid flows Symmetric hyperbolic and thermodynamically compatible(SHTC)systems Hyperbolic systems with curl involutions High-order ADER discontinuous Galerkin(DG)schemes with subcell finite-volume limiter Conservative form of hyperbolic models

来源：评论

学校读者我要写书评

暂无评论

A Prototype Preceding Vehicle Identification System Development and Field Evaluation 10

A Prototype Preceding Vehicle Identification System Developm...

引用

10th International Conference on Vehicle Technology and Intelligent Transport Systems, VEHITS 2024

作者： Mu, Zeyu Tu, Guancheng Shi, Austin Yang, Kun Sun, Yixin Shen, Cong Park, B. Brian Department of Systems & Information Engineering University of Virginia CharlottesvilleVA United States Department of Computer Science University of Virginia CharlottesvilleVA United States Department of Electrical & Computer Engineering University of Virginia CharlottesvilleVA United States Departments of Computer Science and Electrical & Computer Engineering University of Virginia CharlottesvilleVA United States Departments of Civil & Environmental Engineering and Systems & Information Engineering University of Virginia CharlottesvilleVA United States

ISBN: (纸本)9789897587030

Preceding vehicle identification is crucial for establishing cooperative platooning. This paper presents the development of a prototype preceding vehicle identification system (PVIS) and its field evaluation for the assessment of commercial viability. We designed and assembled a prototype consisting of a processing unit (Jetson Nano board), a communication device (Wi-Fi dongle), a GPS unit, and a distance measurement sensor (Terabee sensor). The Jetson Nano integrates the SparkFun GPS-RTK-SMA unit, the Terabee time-of-flight sensor, and the Wi-Fi dongle. The PVIS prototype in the ego vehicle measures the distance to its preceding vehicle and receives the GPS data from potential preceding vehicles with the PVIS prototypes. With these, the PVIS in the ego vehicle determines the connectivity of the preceding vehicle. The field evaluation results showed that the prototype PVIS works as designed, and each successful identification takes about 5.3 seconds. However, it was found that the Terabee (time of flight) sensor, at times, did not properly measure distances, likely due to an angle issue caused by the roadway surface and vibration of the vehicle. We discussed how to overcome the challenges identified and enhance the prototype for successful commercialization. Copyright © 2024 by SCITEPRESS - science and Technology Publications, Lda.

关键词： Vehicle to vehicle communications

来源：评论

学校读者我要写书评

暂无评论

Electromagnetic Modeling Strategy Supporting the Fabrication of Inkjet-Printed Chipless RFID Sensors

IEEE Journal on Flexible Electronics

引用

IEEE Journal on Flexible Electronics 2023年第2期2卷 145-152页

作者： Marchi, Giada Zanazzi, Enrico Mulloni, Viviana Donelli, Massimo Lorenzelli, Leandro Trento38123 Italy University of Trento Department of Information Engineering and Computer Science Trento38123 Italy University of Trento Department of Civil Environmental and Mechanical Engineering Trento38123 Italy

Inkjet-printing fabrication techniques can be used for the manufacturing process of chipless radio frequency identification (RFID) sensors. With respect to traditional silicon-based fabrication, flexible electronics induce the possibility to experiment with new substrates and conductive materials. With the development of high-conductive nano-inks, the low sintering temperature required to process them results in compatibility with a large variety of substrates. However, the requirements for the conductivity of the metalized printed patterns are particularly stringent for applications where the working frequency is high, like in the case of common chipless RFID tags working in the frequency range of 3-10 GHz. In chipless RFID systems, the higher the conductivity of the conductive parts, the better the signal response of the tag. However, in printed chipless RFIDs, the conductivity can only give a partial indication of the goodness of the tag. The tag response is dependent on the sheet resistance, which is not only related to the conductivity of the ink, but also to the thickness of the printed pattern. Therefore, if a sufficiently high-conductive nano-ink is selected, the signal efficiency of the printed tag can be further improved by making an evaluation based on the thickness of the patterns. The objective of this study is to propose a strategy for evaluating the signal response efficiency in inkjet-printed chipless tags by means of a first quantification of the average thickness of the printed layer, which is used to feed a finite-element method (FEM) simulation model to give the manufacturer an idea of the optimal number of layers to be printed to have a strong signal response. © 2022 IEEE.

关键词： Substrates

来源：评论

学校读者我要写书评

暂无评论

Using an Evolutionary Algorithm Such as Genetic Algorithm (Ga) to Optimize Linear Equations 5

Using an Evolutionary Algorithm Such as Genetic Algorithm (G...

引用

5th International Conference on Neural Networks and Neurotechnologies, NeuroNT 2024

作者： Wdaa, Abdul Sttar Ismail Al-Barzinji, Shokhan M. Tarish, Hiba A. College of Education for Pure Science University of Anbar Ramadi Iraq College of Computer Science and Information Technology University of Anbar Department of Computer Networks Systems Ramadi Iraq University of Technology Civil Engineering Departments Baghdad Iraq

ISBN: (纸本)9798350363739

An Evolutionary algorithm is a subset of evolutionary computations. Cloning, mutation, recombination, and selection are some of the methods used by the evolutionary algorithm that were inspired by biological evolution. The problem that was addressed in this study is how to find the coefficient values of the variables in linear equations that contain a set of variables, it must equal the right side. After reviewing previous research looking for optimization, genetic algorithm, which will use in this paper, so how are going to use this strategy to solve our problems? Select the best chromosomes from this population. Perform random chromosomal hybridization. Then, within the genes of our chromosomes, produce random mutations. Will send these chromosomes for selection again. The best chromosomes among them are selected again. It crosses and turns again, and the process is repeated until a certain number of generations have passed or find the best option. After applying the genetic algorithm through the Matlab program to the linear equation, which contains six variables, for example, the results were very excellent, as the parameter values were found in a very short time and with high accuracy, completely matching the right side of the equation. © 2024 IEEE.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Queue Modeling for Geospatial Service on Edge-Cloud Architecture

IEEE Networking Letters

引用

IEEE Networking Letters 2025年第1期7卷 36-40页

作者： Franchi, Fabio Graziosi, Fabio Smarra, Francesco Fina, Eleonora Di Università degli Studi dell'Aquila Computer Science and Mathematics Department of Information Engineering L'Aquila67100 Italy Università degli Studi dell'Aquila Construction-Architectural and Environmental Engineering Department of Civil L'Aquila67100 Italy Fondazione Ugo Bordoni Rome00161 Italy

The exponential growth and complexity of geospatial data necessitate innovative management strategies to address the increasing computational demands of Geographical information System (GIS) services. GIS is connected to the social context, and its use as a decision-support tool is gaining broader acceptance with the need to ensure high Quality of Service (QoS). While cloud computing offers new capabilities for GIS, the physical distance between cloud infrastructure and end-users often leads to high network latency, compromising QoS. Multi-Access Edge Computing (MEC) emerges as a promising solution to limit latency and enhance system performance, particularly for real-time and multi-device applications. However, integrating GIS services into edge-cloud architectures presents significant challenges in terms of task scheduling and service placement. This letter proposes a queueing theory-based model designed to optimize the performance of GIS workloads within edge-cloud architectures. The model, based on a closed Jackson network, is designed to assist in the efficient design and deployment of edge systems that meet QoS and Service Level Agreement (SLA) requirements. The proposed framework is validated through a real-world case study, with performance metrics such as throughput and response time evaluated to ensure optimal system sizing and performance. The results underscore the potential of this approach for designing scalable and efficient edge-cloud architectures tailored to geospatial services. © 2019 IEEE.

关键词： Cloud computing architecture

来源：评论

学校读者我要写书评

暂无评论

Voting Classifier and Metaheuristic Optimization for Network Intrusion Detection

引用

computers, Materials & Continua 2023年第2期74卷 3183-3198页

作者： Doaa Sami Khafaga Faten Khalid Karim Abdelaziz A.Abdelhamid El-Sayed M.El-kenawy Hend K.Alkahtani Nima Khodadadi Mohammed Hadwan Abdelhameed Ibrahim Department of Computer Sciences College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Department of Computer Science Faculty of Computer and Information SciencesAin Shams UniversityCairo11566Egypt Department of Computer Science College of Computing and Information TechnologyShaqra University11961Saudi Arabia Department of Communications and Electronics Delta Higher Institute of Engineering and TechnologyMansoura35111Egypt The Department of Civil and Environmental Engineering Florida International UniversityMiamiFLUSA Department of Information Technology College of ComputerQassim UniversityBuraydah51452Saudi Arabia Computer Engineering and Control Systems Department Faculty of EngineeringMansoura UniversityMansoura35516Egypt

Managing physical objects in the network’s periphery is made possible by the Internet of Things(IoT),revolutionizing human *** attacks and unauthorized access are possible with these IoT devices,which exchange data to enable remote *** attacks are often detected using intrusion detection methodologies,although these systems’effectiveness and accuracy are *** paper proposes a new voting classifier composed of an ensemble of machine learning models trained and optimized using metaheuristic *** employed metaheuristic optimizer is a new version of the whale optimization algorithm(WOA),which is guided by the dipper throated optimizer(DTO)to improve the exploration process of the traditionalWOA *** proposed voting classifier categorizes the network intrusions robustly and *** assess the proposed approach,a dataset created from IoT devices is employed to record the efficiency of the proposed algorithm for binary attack *** dataset records are balanced using the locality-sensitive hashing(LSH)and Synthetic Minority Oversampling Technique(SMOTE).The evaluation of the achieved results is performed in terms of statistical analysis and visual plots to prove the proposed approach’s effectiveness,stability,and *** achieved results confirmed the superiority of the proposed algorithm for the task of network intrusion detection.

关键词： Voting classifier whale optimization algorithm dipper throated optimization intrusion detection internet-of-things

来源：评论

学校读者我要写书评

暂无评论

Machine Learning for Damage Classification, Risk Mitigation and Post-earthquake Management 7th

Machine Learning for Damage Classification, Risk Mitigation ...

引用

7th International Conference on Earthquake engineering and Seismology, ICEES 2023

作者： Di Michele, F. Giannopoulou, O. Stagnini, E. Pera, D. Rubino, B. Aloisio, R. Askan, A. Marcati, P. Milan Italy L’Aquila Italy Department of Information Engineering Computer Science and Mathematics University of L’Aquila L’Aquila Italy Departments of Civil Engineering and Earthquake Studies Middle East Technical University Ankara Turkey Assergi Italy

ISBN: (纸本)9783031573569

In recent years, we have witnessed a steady increase in the amount of data collected and made available to the scientific community. Simultaneously, Artificial Intelligence (AI) shows great potential in transforming data to knowledge offering increasingly more accurate tools to analyze, interpret and extract information from the data. In this context, data-driven approaches, in the fields of seismology, geophysics, and earthquake engineering show enormous promise. In this work a Machine Learning (ML) study is presented, based on a dataset containing around 3000 buildings damaged by the 2009 L’Aquila earthquake. This event was the first in a series of strong earthquakes that hit central Italy, resulting in many casualties, and having enormous economic and social impact. Each building in the dataset is described by 22 characteristics. Among them the damage level, divided, into six classes, from D0, corresponding to no damage, to D5, corresponding to heavy damage or collapse. We employ a Random Forest based algorithm to predict the level of damage accounting for different combinations of damage levels. As a case study we consider the following binary target variables-D0-D1-D2-D3 (no to medium damage) and D4-D5 (serious to heavy damage)-D0-D1 (no to light damage) and D2-D3-D4-D5 (moderate to heavy damage)-D0-D1-D02 (light to moderate damage) and D3-D4-D5 (medium to heavy damage). © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

PathoLM: Identifying pathogenicity from the DNA sequence through the Genome Foundation Model 19

PathoLM: Identifying pathogenicity from the DNA sequence thr...

引用

19th Conference on Machine Learning in Computational Biology Meeting, MLCB 2024

作者： Dip, Sajib Acharjee Shuvo, Uddip Acharjee Chau, Tran Song, Haoqiu Choi, Petra Wang, Xuan Zhang, Liqing Department of Computer Science Virginia Tech United States Institute of Information Technology Dhaka University Bangladesh Virginia Tech United States Department of Civil and Environmental Engineering Virginia Tech United States

Pathogen identification is pivotal in diagnosing, treating, and preventing diseases, crucial for controlling infections and safeguarding public health. Traditional alignment-based methods, though widely used, are computationally intense and reliant on extensive reference databases, often failing to detect novel pathogens due to their low sensitivity and specificity. Similarly, conventional machine learning techniques, while promising, require large annotated datasets and extensive feature engineering and are prone to overfitting. Addressing these challenges, we introduce PathoLM, a cutting-edge pathogen language model optimized for the identification of pathogenicity in bacterial and viral sequences. Leveraging the strengths of pre-trained DNA models such as the Nucleotide Transformer, PathoLM requires minimal data for fine-tuning, thereby enhancing pathogen detection capabilities. It effectively captures a broader genomic context, significantly improving the identification of novel and divergent pathogens. We developed a comprehensive data set comprising approximately 30 species of viruses and bacteria, including ESKAPEE pathogens, seven notably virulent bacterial strains resistant to antibiotics. Additionally, we curated a species classification dataset centered specifically on the ESKAPEE group. In comparative assessments, PathoLM dramatically outperforms existing models like DciPatho, demonstrating robust zero-shot and few-shot capabilities. Furthermore, we expanded PathoLM-Sp for ESKAPEE species classification, where it showed superior performance compared to other advanced deep learning methods, despite the complexities of the task. © 2024 Proceedings of Machine Learning Research. All rights reserved.

关键词： Disease control

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：