检索结果-内蒙古大学图书馆

4th International Conference on Intelligent Computing and Human-computer Interaction, ICHCI 2023

作者： Huang, Hongbo Yan, Xiaoxu Zhang, Xin Zhao, Qisheng Zheng, Yaolin Xu, Longfei Computer School Beijing Information Science and Technology University Beijing China Mechanical Electrical Engineering School Beijing Information Science and Technology University Beijing China

ISBN: (纸本)9798350343113

Deep learning based multi-view stereo algorithms have shown great reconstruction performance. We propose a method called DCMVSNet for depth inference from multi-view images. We choose to build a cost volume pyramid in a coarse-to- fine manner that leads to a lightweight network. It has an efficient feature extraction module based on deformable convolutional network. The proposed multi-stage network architecture can fully utilize multi-scale feature maps to achieve better reconstruction results. Experiments on popular datasets demonstrate that the proposed method achieves the exceptional reconstruction performance. © 2023 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

E-CVP-MVSNet: Effective Hierarchical Multi-View 3D Reconstruction Based on Light-Weight Network 5

E-CVP-MVSNet: Effective Hierarchical Multi-View 3D Reconstru...

引用

5th International Conference on Robotics, Intelligent Control and Artificial Intelligence, RICAI 2023

作者： Zhang, Xin Huang, Hongbo Lu, Lu Yan, Xiaoxu Bao, Zhongman Zhao, Qisheng Gong, Meihui Mechanical Electrical Engineering School Beijing Information Science and Technology University Beijing China Computer School Beijing Information Science and Technology University Beijing China

ISBN: (纸本)9798350357950

Multi-view stereo based on cost volume pyramid has achieved good results in depth inference using high-resolution images. However, most approaches require a lot of computing resources, in part because convolutional layers extract redundant features and pay too much attention to some unwanted feature information. In this work, we propose a pyramid network for feature extraction that is both effective and efficient. There are two main mechanisms in the proposed approach. An effective convolution module called SCConv (Spatial and Channel reconstruction Convolution) to reduce duplicate features and encourage representative feature learning. To encode spatial data and processes high-resolution photos more efficiently, Conv2Former is also introduced into the model to enhance feature extraction capabilities. Experiments show that our method can obviously decrease the parameters, complexity and computational costs, thus improve the performance and generative ability remarkably. © 2023 IEEE.

关键词： Attention mechanism Convolution Depth estimation Feature extraction Multi-view stereo

来源：评论

学校读者我要写书评

暂无评论

Development of Programmer Testing and Question Answering APP Based on SpringBoot 2

Development of Programmer Testing and Question Answering APP...

引用

2nd IEEE International Conference on Sensors, Electronics and computer engineering, ICSECE 2024

作者： Xu, Yihan Xie, Nan Chen, Dengren Wu, Wenjie Chen, Weimin College of Computer Science and Technology Zhejiang University of Water Resources and Electric Power Hangzhou China College of Mechanical and Electrical Engineering China Jiliang University Hangzhou Hangzhou China

ISBN: (纸本)9798350373646

With the rapid development of the information age, the field of programmer programming is at the forefront of change. Programmers face the challenge of constantly learning and improving their programming skills, and sometimes they may not be able to balance improving their competitiveness during interviews. Starting from practical needs, this system designs a programmer testing and question brushing APP based on SpringBoot, which can provide rich programming question banks, interview question testing and other functions, help programmers systematically learn and practice programming knowledge, improve coding efficiency and problem-solving ability, and demonstrate excellent competitiveness in interviews. The system is divided into an APP end and a management end. The APP end displays a carousel chart and a list of questions. Users can filter and search according to their needs to find the questions they want. In addition, the APP end has also added online exam and experience sharing functions. The management end displays data reports and can manage app content and users. In this project, the backend adopts the SpringBoot framework, which realizes the functions of processing front-end requests and interacting with the database;In terms of database, MySQL is used to store user information, question data, etc., and Redis is used for data caching to improve system performance;The front-end part adopts the Vue framework for page construction, and combines Uni-app to achieve cross platform application development, providing users with a good user experience. This project combines modern technology, social reality, and user needs, providing programmers with a rich programming question bank, interview question testing, and other functions to help them systematically learn and practice programming knowledge, improve coding efficiency and problem-solving ability, and provide strong support for them to demonstrate excellent competitiveness in interviews. At the same time, it al

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Global Search of Optimal Spacecraft Trajectories using Amortization and Deep Generative Models

arXiv

引用

arXiv 2024年

作者： Beeson, Ryne Li, Anjian Sinha, Amlan Princeton University NJ08540 United States Department of Mechanical and Aerospace Engineering Department of Electrical and Computer Engineering

Preliminary spacecraft trajectory optimization is a parameter dependent global search problem that aims to provide a set of solutions that are of high quality and diverse. In the case of numerical solution, it is dependent on the original optimal control problem, the choice of a control transcription, and the behavior of a gradient based numerical solver. In this paper we formulate the parameterized global search problem as the task of sampling a conditional probability distribution with support on the neighborhoods of local basins of attraction to the high quality solutions. The conditional distribution is learned and represented using deep generative models that allow for prediction of how the local basins change as parameters vary. The approach is benchmarked on a low thrust spacecraft trajectory optimization problem in the circular restricted three-body problem, showing significant speed-up over a simple multi-start method and vanilla machine learning approaches. The paper also provides an in-depth analysis of the multi-modal funnel structure of a low-thrust spacecraft trajectory optimization problem. Copyright © 2024, The Authors. All rights reserved.

关键词： Optimal control systems

来源：评论

学校读者我要写书评

暂无评论

Physical-based training data collection approach for data-driven lithium-ion battery state-of-charge prediction

引用

Energy and AI 2021年第3期5卷 194-200页

作者： Jie Li Will Ziehm Jonathan Kimball Robert Landers Jonghyun Park Department of Mechanical and Aerospace Engineering Missouri University of Science and TechnologyRollaMO 65401USA Department of Electrical and Computer Engineering Missouri University of Science and TechnologyRollaMO 65401USA

Data-Driven approaches for State of Charge(SOC)prediction have been developed considerably in recent ***,determining the appropriate training dataset is still a challenge for model development and validation due to the considerably varieties of lithium-ion batteries in terms of material,types of battery cells,and operation *** work focuses on optimization of the training data set by using simple measurable data sets,which is important for the accuracy of predictions,reduction of training time,and application to online *** is found that a randomly generated data set can be effectively used for the training data set,which is not necessarily the same format as conventional predefined battery testing protocols,such as constant current cycling,Highway Fuel Economy Cycle,and Urban Dynamometer Driving *** randomly generated data can be successfully applied to various dynamic battery operating *** the ML algorithm,XGBoost is used,along with Random Forest,Artificial Neural Network,and a reduced-order physical battery model for *** XGBoost method with the optimal training data set shows excellent performance for SOC prediction with the fastest learning time within 1 s,a short running time of 0.03 s,and accurate results with a 0.358%Mean Absolute Percentage Error,which is outstanding compared to other Data-Driven approaches and the physics-based model.

关键词： Machine learning Random signal Battery soc Dynamic current Battery modeling Estimation XGBoost

来源：评论

学校读者我要写书评

暂无评论

Design and Analysis of an Actuator with High Power Density Permanent Magnet motor for Bipedal Robots

Design and Analysis of an Actuator with High Power Density P...

引用

2024 International Conference on electrical Machines, ICEM 2024

作者： He, Tianran Leng, Xiaokun Sun, Yangyu Wu, Yucong School of Electronics and Information Engineering Tongji University Shanghai China School of Computer Science and Technology Harbin Institute of Technology Harbin China University of Sheffield Department of Electrical and Electronic Engineering Sheffield United Kingdom Southern University of Science and Technology Department of Mechanical and Energy Engineering Shenzhen China

ISBN: (纸本)9798350370607

Recently, bipedal robots that utilize electric rotary actuators have attracted much attention since they can walk on any kind of path or terrain and can jump or step over obstacles. In addition, bipedal robots are designed to pick up and put down various types of heavy objects for industrial applications. To achieve the specified functions, the electric rotary actuators, as the core driving unit of bipedal robots, need to possess characteristics of high power density and large output torque. This paper presents the design of an electric rotary actuator based on an outer rotor permanent magnet motor for bipedal robots. The proposed electric rotary actuator includes a 36-slot/42-pole permanent magnet motor, a 10:1 planetary gearbox, and an integrated gear shaft. The electromagnetic performances and overload capability of the proposed 36-slot/42-pole outer rotor permanent magnet motor are analyzed by the finite element method. Finally, a prototype of the proposed electric rotary actuator is tested by a customed high-torque testing bench. The experimental results are in good agreement with the finite element analysis results. © 2024 IEEE.

关键词： Permanent magnets

来源：评论

学校读者我要写书评

暂无评论

ELASTIC CONSTANTS INVERSION OF ZIRCONIUM TUBE BASED ON THE LONGITUDINAL MODE OF GUIDED WAVE 18

ELASTIC CONSTANTS INVERSION OF ZIRCONIUM TUBE BASED ON THE L...

引用

18th Symposium on Piezoelectricity, Acoustic Waves, and Device Applications, SPAWDA 2024

作者： Ji, Haoshuai Liu, Hongye Sun, Maoxun Chen, Xiaorong Shen, Liang Qiu, Chongpeng Tian, Zhenhua School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai200093 China Department of Mechanical Engineering Virginia Polytechnic Institute State University 1145 Perry Street BlacksburgVA24061 United States

ISBN: (纸本)9798331523732

Estimating elastic constants in materials is markedly important in engineering and technology. Ultrasonic guided wave non-destructive testing provides an effective means for accurately assessing material properties. This paper presents a method for inverting the elastic constants of hollow pipes using the longitudinal mode of ultrasonic guided waves combined with a genetic algorithm (GA). The Legendre orthogonal polynomial expansion (LOPE) method serves as the forward model for analyzing the dispersion curves of axisymmetric guided wave modes, while the GA performs the inversion. A carefully designed finite element (FE) model generates pseudo-experimental data, which is transformed into the frequency-wave number domain for input into the GA. The comparison between simulation results and theoretical analysis confirms the validity of the FE model and the accuracy of the detected modes. The results demonstrate that the proposed method effectively inverts the elastic constants of isotropic hollow cylindrical structures, with an inversion error of less than 2% based on pseudo-experimental data. © 2024 IEEE.

关键词： Nondestructive examination

来源：评论

学校读者我要写书评

暂无评论

Research on Optimization Methods for Industrial Model Retrieval 5

Research on Optimization Methods for Industrial Model Retrie...

引用

5th International Conference on Natural Language Processing, ICNLP 2023

作者： Wang, Peng Hu, Chunhui Xi'an University of Posts and Telecommunications School of Computer Science Xi'an China Xi'an University of Electronic Science and Technology School of Electrical and Mechanical Engineering Xi'an China

ISBN: (纸本)9798350302219

In the retrieval process of industrial models, the traditional database retrieval can no longer meet their needs in terms of efficiency and precision because of their multi-source heterogeneous, complex types and large information scale. This paper optimizes the Elasticsearch search engine in three aspects: the underlying index of the search engine, keyword search and sorting algorithm;and verifies the feasibility of the method through experiments. © 2023 IEEE.

关键词： Search engines

来源：评论

学校读者我要写书评

暂无评论

Tangentially Aligned Integrated Gradients for User-Friendly Explanations 32

Tangentially Aligned Integrated Gradients for User-Friendly ...

引用

32nd Irish Conference on Artificial Intelligence and Cognitive science, AICS 2024

作者： Simpson, Lachlan Costanza, Federico Millar, Kyle Cheng, Adriel Lim, Cheng-Chew Chew, Hong Gunn School of Electrical and Mechanical Engineering The University of Adelaide Australia School of Computer and Mathematical Sciences The University of Adelaide Australia Information Sciences Division Defence Science and Technology Group Australia

Integrated gradients is prevalent within machine learning to address the black-box problem of neural networks. The explanations given by integrated gradients depend on a choice of base-point. The choice of base-point is not a priori obvious and can lead to drastically different explanations. There is a longstanding hypothesis that data lies on a low dimensional Riemannian manifold. The quality of explanations on a manifold can be measured by the extent to which an explanation for a point lies in its tangent space. In this work, we propose that the base-point should be chosen such that it maximises the tangential alignment of the explanation. We formalise the notion of tangential alignment and provide theoretical conditions under which a base-point choice will provide explanations lying in the tangent space. We demonstrate how to approximate the optimal base-point on several well-known image classification datasets. Furthermore, we compare the optimal base-point choice with common base-points and three gradient explainability models. © 2024 Copyright for this paper by its authors.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Deployment of Large Language Models to Control Mobile Robots at the Edge

arXiv

引用

arXiv 2024年

作者： Sikorski, Pascal Schrader, Leendert Yu, Kaleb Billadeau, Lucy Meenakshi, Jinka Mutharasan, Naveena Esposito, Flavio AliAkbarpour, Hadi Babaiasl, Madi Computer Science Department Aerospace & Mechanical Engineering Department Science Department Robotics Aerospace & Mechanical Engineering Department Saint Louis University Saint Louis United States

This paper investigates the possibility of intuitive human-robot interaction through the application of Natural Language Processing (NLP) and Large Language Models (LLMs) in mobile robotics. This work aims to explore the feasibility of using these technologies for edge-based deployment, where traditional cloud dependencies are eliminated. The study specifically contrasts the performance of GPT-4-Turbo, which requires cloud connectivity, with an offline-capable, quantized version of LLaMA 2 (LLaMA 2-7B.Q5 K M). These results show that GPT-4-Turbo delivers superior performance in interpreting and executing complex commands accurately, whereas LLaMA 2 exhibits significant limitations in consistency and reliability of command execution. Communication between the control computer and the mobile robot is established via a Raspberry Pi Pico W, which wirelessly receives commands from the computer without internet dependency and transmits them through a wired connection to the robot’s Arduino controller. This study highlights the potential and challenges of implementing LLMs and NLP at the edge, providing groundwork for future research into fully autonomous and network-independent robotic systems. For video demonstrations and source code, please refer to: https://***/MobileRobotGPT4LLaMA2024. © 2024, CC BY.

关键词： Human robot interaction

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：