检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Gupta, Hrishikesh Thalhammer, Stefan Weibel, Jean-Baptiste Haberl, Alexander Vincze, Markus Vision for Robotics Laboratory Automation and Control Institute TU Wien Austria Industrial Engineering Department UAS Technikum Vienna TU Wien Austria

Transparent objects are ubiquitous in daily life, making their perception and robotics manipulation important. However, they present a major challenge due to their distinct refractive and reflective properties when it comes to accurately estimating the 6D pose. To solve this, we present ReFlow6D, a novel method for transparent object 6D pose estimation that harnesses the refractive-intermediate representation. Unlike conventional approaches, our method leverages a feature space impervious to changes in RGB image space and independent of depth information. Drawing inspiration from image matting, we model the deformation of the light path through transparent objects, yielding a unique object-specific intermediate representation guided by light refraction that is independent of the environment in which objects are observed. By integrating these intermediate features into the pose estimation network, we show that ReFlow6D achieves precise 6D pose estimation of transparent objects, using only RGB images as input. Our method further introduces a novel transparent object compositing loss, fostering the generation of superior refractive-intermediate features. Empirical evaluations show that our approach significantly outperforms state-of-the-art methods on TOD and Trans32K-6D datasets. Robot grasping experiments further demonstrate that ReFlow6D’s pose estimation accuracy effectively translates to real-world robotics task. The source code is available at: https://***/StoicGilgamesh/ReFlow6D and https://***/StoicGilgamesh/matting *** Codes 68T45 © 2024, CC BY-NC-SA.

关键词： Light refraction

来源：评论

学校读者我要写书评

暂无评论

Reviewing Fault Diagnosis Methods in Electric Drives: Power Subsystem and Electrical Machine 24

Reviewing Fault Diagnosis Methods in Electric Drives: Power ...

引用

24th IEEE International Conference of Young Professionals in Electron Devices and Materials, EDM 2023

作者： Demidova, Galina Chen, Hao Li, Yibin Lukichev, Dmitry Anuchin, Alecksey Itmo University Faculty of Control System and Robotics Saint-Petersburg Russia School of Electrical and Power Engineering China University of Mining and Technology Xuzhou China Institute of Marine Science and Technology Shandong University Shandong China Moscow Power Engineering Institute Department of Electric Drives Moscow Russia

ISBN: (纸本)9798350336870

This paper provides a review of popular fault diagnosis methods in the power subsystem and electrical machine of electric drives. The review aims to explore the different approaches used to detect faults in these components and the most common types of faults encountered in electric drives. The paper first provides an overview of the power subsystem and electrical machine and their respective functions within the electric drive system. It then proceeds to discuss the different types of faults that can occur in these components, including overvoltage, undervoltage, overcurrent, and short circuits. The review evaluates the various fault diagnosis methods available for these components, including model-based, signal-based, and artificial intelligence-based approaches. Model-based approaches use mathematical models of the system to detect and diagnose faults. Signal-based approaches rely on the analysis of measured signals to detect changes in system behavior that indicate a fault. Artificial intelligence-based approaches use machine learning algorithms to identify patterns in data and detect faults. Model-based approaches are highly accurate but can be computationally intensive. Signal-based approaches are less computationally intensive but may be less accurate in certain situations. Artificial intelligence-based approaches are versatile but require significant amounts of data for training. This review provides a comprehensive overview of the current state-of-the-art in fault diagnosis methods for the power subsystem and electrical machine of electric drives. The review concludes by suggesting potential areas for future research to further improve the accuracy and efficiency of fault diagnosis methods for electric drives. © 2023 IEEE.

关键词： Failure analysis

来源：评论

学校读者我要写书评

暂无评论

Design and Implementation of Pick and Place Manipulation System for Industrial Automation

Design and Implementation of Pick and Place Manipulation Sys...

引用

2021 International Conference on Artificial Intelligence and Mechatronics Systems, AIMS 2021

作者： Naeem, Mehreen Aslam, Shoaib Suhaib, Muhammad Gul, Seemab Murtaza, Zeeshan Khan, Muhammad Jawad National University of Science and Technology Department of Robotics and Intelliegnt Machine Engineering Pakistan University of Engineering and Technology Lahore Department of Mechatronics and Control Engineering Pakistan

ISBN: (纸本)9781665424820

This work presents an autonomous pick and pack system for industrial applications consisting of a 4-DOF (degree-of-freedom) robotic manipulator and the vision perception. The system is capable to detect and capture products/materials from a production line and place them in a packaging box one at a time. The final placement of the object in the box is carried out with the desired orientation. The adopted configuration of the manipulator is RRPR equipped with an electromagnetic end-effector. Inverse kinematics approach is used to determine the end-effector motion sequence in order to reach the desired locations within the workspace. The manipulator is equipped with Feetech SC servo motor with built-in PID controller at each joint to achieve precise motion of joint angles. An RGB camera is mounted right above the conveyer belt to capture the region of interest in the environment. Captured image is further preprocessed by image processing techniques. Region properties is used to extract the orientation of the object. Arduino Mega 2560 is used as a control board for motors and an interface with the PC. A graphical user interface (GUI) is developed in MATLAB to monitor and control the end-to-end pick and pack system featuring both manual and automatic modes. © 2021 IEEE.

关键词： End effectors

来源：评论

学校读者我要写书评

暂无评论

A Practical Algorithm for Multirotor System Design and Evaluation

A Practical Algorithm for Multirotor System Design and Evalu...

引用

RSI/ISM International Conference on robotics and Mechatronics (ICROM)

作者： Arshia Rezaei Faezeh Vahedi Amin Talaeizadeh Aria Alasty Department of Mechanical Engineering Advanced Research Lab for Control and Agricultural Robotics (Sharif AgRoLab) Sharif University of Technology

Multirotors have attracted much attention in recent years. During the design phase, both designers and users are often curious about whether an assembled multirotor can meet their expectations, including hover endurance, system efficiency, maximum load capacity, maximum pitch angle, and maximum flight distance. Traditionally, multirotor performance evaluation involves multiple flight tests or relying on experience, which can be time-consuming and expensive. To solve this challenge, this algorithm is proposed for extensive offline design and evaluation. This algorithm primarily focuses on performance metrics linked to the propulsion system, encompassing components such as motors, propellers, electronic speed controllers, and batteries. In the initial stage of our research, we establish models for these propulsion system components. Notably, the input for these models comprises technical specifications provided by manufacturers. Using these models and their interrelationships, equations were formulated to describe different performance indicators. Then these equations were solved to perform performance evaluation. The effectiveness of this evaluation method is shown through an example developed by a powerful and user-friendly software that enables users to access the performance evaluation approach presented in this article.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Keypoints Estimation on Partial Point Clouds

Keypoints Estimation on Partial Point Clouds

引用

International Conference on robotics and Artificial Intelligence (ICRAI)

作者： Maryam Saleem Mohammad Zohaib Syed Abbas Zilqurnain Naqvi Department of Mechatronics & Control Engineering Human Centered Robotics Lab UET Lahore Pakistan SIGMA Clermont University of Clermont Auvergne Aubiere France

ISBN: (数字)9798350355703

ISBN: (纸本)9798350355710

Estimating 3D keypoints of an object has achieved considerable interest, especially in computer graphics and ma-chine vision. The complexity of the problem increases when the object is noisy, decimated, or partially occluded. To this end, we present an approach to estimate 3D keypoints from the partial PCDs such that they also highlight the object's missing parts. It is based on two modules: shape completion and keypoints estimation. The first module computes the object's missing parts and generates a dense shape which is used by the second module to estimate semantically and geometrically consistent keypoints. The approach also estimates confidence scores that allow the selection of valid keypoint considering the object's structure. We evaluate the proposed method for four different backbones and observe that we get better results using PointNet. Furthermore, the presented results show that the Percentage of Correct Keypoints (PCK) and Dual Alignment Score (DAS) for the keypoints estimated from the dense shapes (36.61 and 76.29) are 23.51 % and 12.14% higher, respectively, compared to those estimated from the partial objects (13.10 and 64.15).

关键词： Point cloud compression Industries Three-dimensional displays Accuracy Shape Estimation Computer graphics Complexity theory Noise measurement robotics and automation

来源：评论

学校读者我要写书评

暂无评论

ZS6D: Zero-shot 6D Object Pose Estimation using Vision Transformers

ZS6D: Zero-shot 6D Object Pose Estimation using Vision Trans...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Philipp Ausserlechner David Haberger Stefan Thalhammer Jean-Baptiste Weibel Markus Vincze Vision for Robotics Laboratory Automation and Control Institute TU Wien Austria Industrial Engineering Department University of Applied Sciences Technikum Vienna Austria

ISBN: (数字)9798350384574

ISBN: (纸本)9798350384581

As robotic systems increasingly encounter complex and unconstrained real-world scenarios, there is a demand to recognize diverse objects. The state-of-the-art 6D object pose estimation methods rely on object-specific training and therefore do not generalize to unseen objects. Recent novel object pose estimation methods are solving this issue using task-specific fine-tuned CNNs for deep template matching. This adaptation for pose estimation still requires expensive data rendering and training procedures. MegaPose for example is trained on a dataset consisting of two million images showing 20,000 different objects to reach such generalization capabilities. To overcome this shortcoming we introduce ZS6D, for zero-shot novel object 6D pose estimation. Visual descriptors, extracted using pre-trained Vision Transformers (ViT), are used for matching rendered templates against query images of objects and for establishing local correspondences. These local correspondences enable deriving geometric correspondences and are used for estimating the object's 6D pose with RANSAC- based PnP. This approach showcases that the image descriptors extracted by pre-trained ViTs are well-suited to achieve a notable improvement over two state-of-the-art novel object 6D pose estimation methods, without the need for task-specific fine-tuning. Experiments are performed on LMO, YCBV, and TLESS. In comparison to MegaPose, we improve the Average Recall on all three datasets and compared to OSOP we improve on two datasets. The code is available at https://***/PhilippAuss/ZS6D.

关键词： Training Computer vision Visualization Codes Computational modeling Pose estimation Transformers

来源：评论

学校读者我要写书评

暂无评论

Graph-Based Collision Avoidance Algorithm Among Swarm Agents

Graph-Based Collision Avoidance Algorithm Among Swarm Agents

引用

International Symposium on Electrical and Electronics engineering (ISEEE)

作者： Akif Durdu Mesut Tuncer Berat Yıldız Department of Electrical and Electronics Engineering Robotics Automation Control Laboratory (RAC-LAB) Konya Technical University Konya Turkiye Department of Computer Engineering Robotics Automation Control Laboratory (RAC-LAB) Konya Technical University Konya Turkiye Department of Electrical and Electronics Engineering Karamanoglu Mehmetbey University Karaman Turkiye

More than one homogenous or heterogenous type unmanned vehicle can work in a coordinated manner and perform large-scale swarm tasks (firefighting, search and rescue, mapping, and military operations, etc.) efficiently in a shorter time by sharing tasks. Collision of these vehicles is among the most significant problems encountered during their work. The crash of the vehicles causes the vehicles to be out of duty and, accordingly, to the mission’s failure. In this study, Quadrotor-type UAVs used as agents can go to any target point by receiving location, speed, and compass information with GPS and IMU sensors. In the application, the agents’ locations were kept and updated in a list in pairs, similar to the traversing process in the Optimized Bubble Sort Algorithm. The projections of the velocity vectors on the agents’ axis (local) on a single coordinate plane are taken to determine the collision situation between these two agents that are traveling instantaneously. These global velocity vectors, whose projections are taken, are re-projected to the edge formed by these two agents in the graph and then subtracted from each other. Suppose the size of the vector is greater than the distance between two agents obtained by any GPS distance algorithm (Pythagoras, Haversine, etc.). In this case, collision is detected, and the separation process is activated. Separation is the process of advancing one agent in the opposite direction of the other until a minimum safe distance is achieved, where one agent entered by the user will not affect the other. Once the separation is complete, the agent moves to the final destination point.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Data-Efficient System Identification via Lipschitz Neural Networks

arXiv

引用

arXiv 2024年

作者： Wei, Shiqing Krishnamurthy, Prashanth Khorrami, Farshad Control/Robotics Research Laboratory Department of ECE NYU Tandon School of Engineering 5 MetroTech Center BrooklynNY11201 United States

Extracting dynamic models from data is of enormous importance in understanding the properties of unknown systems. In this work, we employ Lipschitz neural networks, a class of neural networks with a prescribed upper bound on their Lipschitz constant, to address the problem of data-efficient nonlinear system identification. Under the (fairly weak) assumption that the unknown system is Lipschitz continuous, we propose a method to estimate the approximation error bound of the trained network and the bound on the difference between the simulated trajectories by the trained models and the true system. Empirical results show that our method outperforms classic fully connected neural networks and Lipschitz regularized networks through simulation studies on three dynamical systems, and the advantage of our method is more noticeable when less data is used for training. © 2024, CC BY.

关键词： Neural network models

来源：评论

学校读者我要写书评

暂无评论

A control Theoretic Study on Omnidirectional MAVs with Minimum Number of Actuators and No Internal Forces at Any Orientation

A Control Theoretic Study on Omnidirectional MAVs with Minim...

引用

IEEE Conference on Decision and control

作者： Ahmed Ali Chiara Gabellieri Antonio Franchi Robotics and Mechatronics Department Electrical Engineering Mathematics and Computer Science (EEMCS) Faculty University of Twente Enschede Netherlands Department of Computer Control and Management Engineering Sapienza University of Rome Rome Italy

ISBN: (数字)9798350316339

ISBN: (纸本)9798350316346

We propose a new multirotor aerial vehicle class of designs composed of a multi-body structure in which a main body is connected by passive joints to links equipped with propellers. We have investigated some instances of such class, some of which are shown to achieve omnidirectionality while having a minimum number of inputs equal to the main body Degrees of Freedom DoF’s, only uni-directional positive thrust propellers, and no internal forces generated at steady state. After dynamics are derived following the Euler-Lagrange approach, an I/O dynamic feedback linearization strategy is then used to show the controllability of any desired pose with stable zero dynamics. We finally verify the developed controller with closed-loop simulations.

关键词： Three-dimensional displays Propellers Dynamics controllability Regulation Steady-state Feedback linearization control theory Closed loop systems Vehicle dynamics

来源：评论

学校读者我要写书评

暂无评论

An Efficient Medical Image Deep Fusion Model Based on Convolutional Neural Networks

引用

Computers, Materials & Continua 2023年第2期74卷 2905-2925页

作者： Walid El-Shafai Noha A.El-Hag Ahmed Sedik Ghada Elbanby Fathi E.Abd El-Samie Naglaa F.Soliman Hussah Nasser AlEisa Mohammed E.Abdel Samea Department of Electronics and Electrical Communications Engineering Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt Security Engineering Laboratory Department of Computer SciencePrince Sultan UniversityRiyadh11586Saudi Arabia Higher Institute of Commercial Science Management Information SystemsEl-Mahala El-KobraEgypt Department of the Robotics and Intelligent Machines Faculty of Artificial IntelligenceKafrelsheikh UniversityKafr El-SheikhEgypt Department of Industrial Electronics and Control Engineering Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt Department of Information Technology College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Department of Computer Sciences College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversityRiyadhSaudi Arabia Medical Imaging and Interventional Radiology National Liver InstituteMenoufia UniversityEgypt

Medical image fusion is considered the best method for obtaining one image with rich details for efficient medical diagnosis and *** learning provides a high performance for several medical image analysis *** paper proposes a deep learning model for the medical image fusion *** model depends on Convolutional Neural Network(CNN).The basic idea of the proposed model is to extract features from both CT and MR ***,an additional process is executed on the extracted *** that,the fused feature map is reconstructed to obtain the resulting fused ***,the quality of the resulting fused image is enhanced by various enhancement techniques such as Histogram Matching(HM),Histogram Equalization(HE),fuzzy technique,fuzzy type,and Contrast Limited Histogram Equalization(CLAHE).The performance of the proposed fusion-based CNN model is measured by various metrics of the fusion and enhancement *** realistic datasets of different modalities and diseases are tested and ***,real datasets are tested in the simulation analysis.

关键词： Image fusion CNN deep learning feature extraction evaluation metrics medical diagnosis

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：