检索结果-内蒙古大学图书馆

8th Conference on Robot Learning, CoRL 2024

作者： Song, Pinhao Li, Pengteng Detry, Renaud KU Leuven Dept. Mechanical Engineering Research Unit Robotics Automation and Mechatronics Belgium KU Leuven Dept. Electrical Engineering Research Unit Processing Speech and Images Belgium AI Thrust Hong Kong

There are two dominant approaches in modern robot grasp planning: dense prediction and sampling-based methods. Dense prediction calculates viable grasps across the robot's view but is limited to predicting a fixed number of grasps per voxel. Sampling-based methods, on the other hand, encode multi-modal grasp distributions, allowing for different grasp approaches at a point. However, these methods rely on a global latent representation, which struggles to represent the entire field of view, resulting in coarse grasps. To address this, we introduce Implicit Grasp Diffusion (IGD), which combines the strengths of both methods by using implicit neural representations to extract detailed local features and sampling grasps from diffusion models conditioned on these features. Evaluations on clutter removal tasks in both simulated and real-world environments show that IGD delivers high accuracy, noise resilience, and multi-modal grasp pose capabilities. Our code is freely available at https://***/detry-lab/public/***. © 2024 Proceedings of Machine Learning Research.

关键词： Robot programming

来源：评论

学校读者我要写书评

暂无评论

Advancing Pedestrian Attribute Classification: A Multi-Head Deep Learning Paradigm 12

Advancing Pedestrian Attribute Classification: A Multi-Head ...

引用

2024 12th International Conference on Systems and Control, ICSC 2024

作者： Bekbolat, Amanali Kurokawa, Syuhei Tanveer, Muhammad Hassan Ali, Md. Hazrat Nazarbayev University SEDS Dept. of Mechanical and Aerospace Engineering Astana Kazakhstan Kyushu University Dept. of Mechanical Engineering Nishi-ku Fukuoka819-0395 Japan Kennesaw State University Dept. of Robotics and Mechatronics Engineering MariettaGA30067 United States

ISBN: (纸本)9798331531812

Evaluating pedestrians is critical when designing new systems for more sophisticated video monitoring and innovative computer vision-based security. Although convolutional neural networks successfully learn different features from images, modeling the pedestrians to the details needed for such tasks remains an issue. In this paper, we present a novel approach based on the multi-head deep learning model to identify the pedestrian attributes, which has merits in this method. Thus, the proposed approach yields better accuracy, robustness, and efficiency in comparison to state-of-the-art methods on a novel database consisting of pedestrian images. We also incorporate further pre-processing steps whereby data augmentation, transfer learning, and model ensembling are additional methods that are used to improve the generality and applicability of the approach proposed in this study. This framework empirically achieves promising accuracy and outperforms existing approaches, proving the usefulness and robustness of our idea in terms of pedestrian attribute recognition. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Independent Optimization for Robot Path Planning and Dynamic Obstacle Avoidance 19

Independent Optimization for Robot Path Planning and Dynamic...

引用

19th IEEE International Conference on Smart Communities: Improving Quality of Life Using ICT, IoT and AI, HONET 2022

作者： Hall, Terrance Johnson, Christopher Swales, Brighton Koduru, Charles Tanveer, M. Hassan Kennesaw State University Dept. of Robotics and Mechatronics Engineering Marietta United States

ISBN: (纸本)9781665461979

Autonomous robots can be assigned with various tasks such as moving payload, analyzing terrain, and capturing data in an environment. For an Autonomous Mobile Robot (AMR) to execute such tasks the robot (Hussarion ROSbot) will require efficient algorithms and techniques to reference its current location. The robot is relative to surrounding obstacles in its predetermined path. The conducted research study explains the coordinated method used to successfully allow a robot to identify its position in the environment (Gazebo Simulation) and avoid obstructions with increasing velocity - contingent on nearby surroundings. The results show multiple robots individually tasked with distinct roles, while incorporating an obstacle avoidance function used to avoid both static and dynamic obstacles. Such results can be used in the applications of a high-capacity warehouse environments. © 2022 IEEE.

关键词： Multipurpose robots

来源：评论

学校读者我要写书评

暂无评论

Design and Analysis of a Novel Hybrid Grasping Mode Microgripper for High Precision Tasks 9

Design and Analysis of a Novel Hybrid Grasping Mode Microgri...

引用

9th IEEE International Women in engineering (WIE) Conference on Electrical and Computer engineering, WIECON-ECE 2023

作者： Das, Tilok Kumar Shahrear, Rifat Das, Arpita Shirinzadeh, Bijan Dept. of Mechanical Engineering Chattogram Bangladesh Temple University Dept. of Electrical and Computer Engineering PhiladelphiaPA United States Monash University Robotics and Mechatronics Research Laboratory Dept. of Mechanical and Aerospace Engineering MelbourneVIC Australia

ISBN: (纸本)9798350319651

This paper represents a novel piezoelectric actuated flexure-based compliant microgripper which can perform both parallel and angular grasping regardless of the geometric surface of the objects. Two compliant bridge-type mechanisms along with two symmetrical microgripper jaws at the end of the two side arms are accommodated to construct the microgripper. The two side arms of the microgripper act as lever-type amplification mechanisms. Simultaneous actuation of two bridge-type mechanisms is needed to perform linearized parallel grasping. The symmetric design is adopted to maintain better stability. The Finite Element Analysis (FEA) has been performed to evaluate the performance of the microgripper. FEA shows that each jaw of the microgripper can achieve a displacement of 141.72 μm resulting in a total displacement of 283.44 μm with an amplification ratio of 10.7 and 0.715% parasitic motion in y-direction of the X -axis displacement during parallel grasping. Also, each jaw can achieve a displacement of 283.88 μm which results in a total displacement of 567.76 μm with an amplification ratio of 21.5 and 5.17 mrad angular rotation about the Z-axis during angular grasping. The FEA suggests that the microgripper is capable of both parallel and angular grasping/releasing tasks. © 2023 IEEE.

关键词： Piezoelectric actuators

来源：评论

学校读者我要写书评

暂无评论

Guidance Lanes and Driving Algorithms Based on Computer Vision for Speed Control of Autonomous Mobile Robot

引用

Transactions of the Korean Institute of Electrical Engineers 2024年第2期73卷 433-438页

作者： Kim, Jung-Ju Choi, Jae-Hoon Kim, Dong-Jin Choi, Sang-Yule Dept. of Robotics Hoseo University Korea Republic of Dept. of Mechatronics Engineering Induk University Korea Republic of

Conventional mobile robots control driving speed using various sensors data. However, in this paper, we present the speed control method for mobile robot using a single vision sensor and a guidance lane. The proposed guidance lane contains the speed control information of the mobile robot. In addition, installation is simple and economical without the need for embedding separate guidance devices. The speed control accuracy of the mobile robot was obtained through experiments. The experimental results demonstrated that the mobile robot system proposed in this paper is capable of accurate speed control with an accuracy of 97.30 %. Therefore, it is expected to make an important contribution to expanding the scope of use of mobile robots to various application fields. © 2024 Korean Institute of Electrical Engineers. All rights reserved.

关键词： Computer vision

来源：评论

学校读者我要写书评

暂无评论

Optimal Design of Parallel Elastic Actuators for Minimizing Required Torque and Power: Comparison with Series Elastic Actuators

Optimal Design of Parallel Elastic Actuators for Minimizing ...

引用

Annual Conference of Industrial Electronics Society

作者： Kyeongsik Shin Sehoon Oh Dept. Robotics and Mechatronics Engineering DGIST Daegu Republic of Korea

ISBN: (数字)9781665464543

ISBN: (纸本)9781665464550

This paper proposes an optimal design methodology for Parallel Elastic Actuators (PEAs) that minimizes torque and power consumption for tasks with external conditions. Furthermore, the paper advocates for selecting an actuator based on a comparison of the dynamic characteristics of PEAs and Series Elastic Actuators (SEAs) for a specific task. However, the design of actuators may need to vary depending on the task and application, but the criteria for this are often unclear. To address the ambiguous design criteria, we analyze the dynamic characteristics of PEAs and SEAs under external conditions and utilize them for optimal design. By formulating cost functions for various tasks, we identify the most suitable actuator design for each task. This study establishes criteria for designing PEAs for tasks with external conditions and proposes a general framework for PEA design optimization.

关键词： Industrial electronics Actuators Torque Power demand Gears Design methodology Cost function Numerical models Springs Design optimization

来源：评论

学校读者我要写书评

暂无评论

Development of Smart Orthosis with Real-time Corrective Feedback 17

Development of Smart Orthosis with Real-time Corrective Feed...

引用

17th International Conference on engineering of Modern Electric Systems, EMES 2023

作者： Oarcea, Alexandru Popister, Florin Ciudin, Paul Salatioan, Michael Christian The Politechnic University of Timisoaora Dept. of Mechatronics Timisoara Romania Technical University of Cluj-Napoca Dep. of Design Engineering and Robotics Cluj-Napoca Romania

ISBN: (纸本)9798350310634

In this paper we analyze the development process of a smart knee orthosis that can monitor the bearer's movements and provide real-time feedback to aid the bearer's proper recovery after an injury. The development process includes the 3D scanning of the patients knee joint, the 3D modeling of the orthosis, the 3D printing of the orthosis, the integration of IMU (Inertial Measurement Unit) sensors, the integration of the microcontroller and the implementation of the data acquisition and control. The implementation with the use of 3D printing parts will guarantee the best fit on the wearer's foot, offering increased comfort. © 2023 IEEE.

关键词： Microcontrollers

来源：评论

学校读者我要写书评

暂无评论

FFT-UAVNet: FFT Based Human Action Recognition for Drone Surveillance System

FFT-UAVNet: FFT Based Human Action Recognition for Drone Sur...

引用

Sustainable Technologies for Industry 4.0 (STI), International Conference on

作者： Abdul Monaf Chowdhury Ahsan Imran Md Mehedi Hasan Dept. of Robotics and Mechatronics Engineering University of Dhaka Bangladesh

Unmanned aerial vehicles (UAVs) have emerged as a transformative technology for human action recognition, providing a birds-eye view and unlocking new possibilities for precise and comprehensive support in surveillance systems. While substantial advances in ground-based human action recognition have been achieved, the unique characteristics of UAV footage present new challenges that require tailored solutions. Specifically, the reduced scale of humans in aerial perspectives necessitates the development of specialised models to accurately recognize and interpret human actions. Our research focuses on modifying the well-established C3D model and incorporating Fast Fourier Transform (FFT)-based object disentanglement (FO) and space-time attention (FA) mechanisms. By leveraging the power of FFT, our model effectively disentangles the human actors from the background and captures the spatio-temporal dynamics of human actions in UAV footage, enhancing the discriminative capabilities and enabling accurate action recognition. Through extensive experimentation on a subset of the UAV-Human dataset, our proposed FFT-UAVNet (m-C3D+FO&FA+FC) model demonstrates remarkable improvements in performance. We achieve a Top-1 accuracy of 64.86% and a Top-3 accuracy of 83.37%, surpassing the results obtained by the standard C3D and X3D methods, which achieve only a Top-1 accuracy of 28.05% and 31.33%, respectively. These findings underscore the efficacy of our approach and emphasize the significance of the proposed model for UAV datasets in maximizing the potential of UAV-based human action recognition.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Prescribed Performance Sliding Mode Control and Estimation for Perturbed Double-Pendulum Cranes

Prescribed Performance Sliding Mode Control and Estimation f...

引用

Annual Conference of Industrial Electronics Society

作者： Ahmad Bala Alhassan Muhammad Auwal Shehu Ton Duc Do Dept. of Robotics and Mechatronics School of Engineering and Digital Science (SEDS) Nazarbayev University Astana Kazakhstan Dept. of Mechatronics Engineering Nigerian Defence Academy Kaduna Nigeria

ISBN: (数字)9781665464543

ISBN: (纸本)9781665464550

Cranes suffer significantly from parameter uncertainties, frictions, unknown dynamics, and disturbances. Besides, it may be necessary to design a state estimator for an effective feedback controller design. Furthermore, the performance quality and precision of model-based controllers and state estimators rely heavily on accurate models, which may be challenging to formulate. In this study, to address the highlighted control problems and also achieve a precise level of bogie tracking error, the sliding mode control (SMC) with a novel prescribed performance function-based sliding surface is designed. Simulations have shown that the proposed control technique achieved pre-specified transient and steady-state quality metrics on the bogie tracking error where the error converged to a pre-defined small vestigial set, with a decay rate no less than the pre-assigned rate, exhibiting overshoots and undershoots less than some small-enough pre-assigned values.

关键词： Measurement Industrial electronics Uncertain systems Cranes Friction Estimation Observers Steady-state Transient analysis Sliding mode control

来源：评论

学校读者我要写书评

暂无评论

Shadow Detection of Building Facades for Energy Efficiency Using YOLOv8 and Segmentation Techniques

Shadow Detection of Building Facades for Energy Efficiency U...

引用

Artificial Intelligence, robotics and Control (AIRC), International Conference on

作者： Nardine Radany Mostafa Abdelaziz Mechatronics and Robotics Mechanical Engineering Dept. The British University in Egypt (BUE) Cairo Egypt

ISBN: (数字)9798350385076

ISBN: (纸本)9798350385083

Effectively utilizing natural light in architectural design has a substantial impact on energy usage in addition to improving aesthetic appeal. This study presents a novel use of shadow detection for energy-saving projects and architects. We use two different AI approaches, segmentation techniques and YOLOv8, to automate recognizing shadows thrown on building facades. This paper starts with a thorough explanation of the importance of shadow analysis in architectural design, highlighting how it can maximize natural lighting and reduce energy use. First, we study how to recognize shadows on building facades using YOLOv8, a state-of-the-art object identification framework. This paper discusses how YOLOv8 was modified for this particular use case. Secondly, we explore Otsu's threshold segmentation method as an alternative approach to shadow detection. The YOLOv8 shadow detection approach is favored for its versatility across diverse input image types, enabling effective shadow detection even in dissimilar images. It efficiently identifies shadows by generating bounding boxes that indicate the degree of shadow accuracy. Otsu's threshold image segmentation is recommended for situations where additional image processing is intended, such as determining shadow areas using methods like background subtraction.

关键词： Image segmentation Renewable energy sources Accuracy Shape Image color analysis Buildings Process 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：