检索结果-内蒙古大学图书馆

Iberian robotics Conference (ROBOT)

作者： Dylan Denizon Sedat Dogru Lino Marques Department of Electrical and Computer Engineering Institute of Systems and Robotics University of Coimbra Coimbra Portugal

ISBN: (数字)9798350376364

ISBN: (纸本)9798350376371

In this work, we aim to improve the performance of an underactuated anthropomorphic tendon driven robotic gripper with two fingers by allowing the stiffness of the finger joints to change as a function of the object shape. An optimization problem requiring that the gripper apply a constant total contact force with a minimum actuation force was formulated using a Genetic Algorithm (GA) approach in order to obtain a set of designs with geometrical parameters such as position and size of various elements of the fingers as well as stiffness of the joints. The optimization process was also repeated for the fixed stiffness case to compare the two approaches, demonstrating that the addition of variable stiffness allowed to achieve the same grasp with less actuation force.

关键词： Torque Shape Friction Force Grasping Grippers Joining processes Robots Tendons Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Multimodal Human Detection Using YOLO and Representation Learning for Robot Perception

Multimodal Human Detection Using YOLO and Representation Lea...

引用

Iberian robotics Conference (ROBOT)

作者： Kennedy O. S. Mota Diogo S. De Oliveira Luís Garrote Cristiano Premebida Department of Electrical and Computer Engineering University of Coimbra Institute of Systems and Robotics Coimbra Portugal

ISBN: (数字)9798350376364

ISBN: (纸本)9798350376371

This work concentrates on the problem of multisensor people detection using YOLO trained on four distinct modalities: depth and intensity LiDAR-maps, RGB, and ‘thermal’ images. RGB cameras, ubiquitous in this application domain, offer great resolution but struggle with adverse lighting conditions resulting in overexposed or underexposed images which then impact negatively on the performance of the algorithms. Thermal (long-wave infrared) cameras are more resilient against varying light conditions and provide complementary textural features, although with lower resolution when compared to RGB cameras. LiDAR sensors, while having a significantly low resolution, contribute to a physically interpretable mapping of the environment providing precise information regarding size/dimension and location of the objects. The main goal of this work is to tackle people detection using deep-models trained on single and multi-modality representations. To support the experimental part this work introduces a new multimodal dataset (called MID-3K). MID-3K allows the development of data fusion strategies by leveraging four modalities (obtained from three distinct exteroceptive sensors mounted on a mobile robot). Leveraging on a single-modality YOLO framework, we propose a multimodal representation learning approach to improve the baseline performance and to capture more relevant features across all input modalities. The evaluation of the proposed detection pipeline is conducted on the MID-3K dataset, where the reported results are grounded on state-of-the-art performance measures. The new dataset is available in a GitHub repository 1 1 MID-3K dataset: https://***/MID-3K/.

关键词： YOLO Representation learning Training Image resolution Laser radar Robot vision systems Thermal sensors Sensor fusion Cameras Software development management

来源：评论

学校读者我要写书评

暂无评论

Exploiting 3D Grids for Indoor SLAM in Featureless Scenarios

Exploiting 3D Grids for Indoor SLAM in Featureless Scenarios

引用

IEEE International Conference on Autonomous Robot systems and Competitions (ICARSC)

作者： Luís Garrote Ulisses Reverendo Urbano J. Nunes Institute of Systems and Robotics Department of Electrical and Computer Engineering University of Coimbra Coimbra Portugal

ISBN: (数字)9798350352344

ISBN: (纸本)9798350352351

Accurate multi-sensor localization is a challenging task in the navigation of AMRs. Precise localization strategies are essential for AMRs to be able to perform with safety their missions in their surrounding environments. This work proposes a novel ROS-based modular 3D grid-based particle filter-based framework that can be used for Simultaneous Localization and Mapping (SLAM) or as a standalone robust localization strategy. The framework uses odometry and 3D LiDAR data as inputs for localization and SLAM. To further improve localization and representation alignment, a pose refinement stage is employed using Levenberg-Marquardt minimization. The refinement stage considers keypoints in the environment to improve localization and uses the raw 3D point cloud for map maintenance. A pyramid-like 3D grid resolution is used to aid the refinement of the representation, improving pose estimates in featureless scenarios. Experimental validation was carried out with data acquired using an in-house platform, in a set of indoor and semi-structured scenarios comprised of critical featureless areas. The obtained results highlight the robustness of the proposed framework in both SLAM and localization tasks. The code (ROS package) is made available in a GitHub repository 1 .

关键词： Location awareness Point cloud compression Three-dimensional displays Simultaneous localization and mapping Navigation Robustness Safety

来源：评论

学校读者我要写书评

暂无评论

Adaptive-extremum seeking control algorithm to seek the odor source using a multi-agent system 41

Adaptive-extremum seeking control algorithm to seek the odor...

引用

第41届中国控制会议

作者： Meh Jabeen Qing-Hao Meng Li-Cheng Jin Sheng Jin Zhen-Yu Deng Institute of Robotics and Autonomous Systems School of Electrical and Information EngineeringTianjin University Department of Software Engineering The University of Lahore Tianjin Fire Science and Technology Research Institute of MEM

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

This study aims to seek the odor source using a multi-agent *** main challenge of the odor source seeking is to search the optimal extremum in the presence of multiple local extrema in a gradient complex *** this,an adaptive-extremum seeking control(adaptive-ESC) algorithm has been *** are navigated to the vicinity of the source based on an estimated gradient using *** the proposed algorithm,both the integral gain and the amplitude of the perturbation signal are adjusted *** local convergence of the algorithm has been proved using averaging *** performance of the proposed algorithm is validated through simulations by employing a multi-agent system.

关键词： Odor source seeking Extremum seeking control Adaption approaches

来源：评论

学校读者我要写书评

暂无评论

Soft Sensors for Industrial Applications: Comparison of Variables Selection Methods and Regression Models

Soft Sensors for Industrial Applications: Comparison of Vari...

引用

2023 International Conference on Control, Automation and Diagnosis, ICCAD 2023

作者： Leiria, Joana Salles, Rodrigo Mendes, Jerome Sousa, Pedro University of Coimbra Institute of Systems and Robotics Department of Electrical and Computer Engineering Coimbra3030-290 Portugal Oncontrol Technologies Rua Cidade Poitiers Coimbra3000-108 Portugal

ISBN: (纸本)9798350347074

The digitalization of industrial environments has enabled the development of tools that make the production process more efficient and safer. In this sense, the Soft Sensor (SS) plays a fundamental role. Through historical data and indirect measurements, it is possible to estimate the value of important variables that are difficult to measure. This paper presents the SS development process: data collection and pre-processing, variable selection, model selection for SS implementation, model training and testing, and performance evaluation. The selection of variables was made with the help of Pearson Correlation, Mutual Information, and fastTracker algorithm techniques. For the implementation of SS have been tested several models: Multiple Linear Regression, Ridge Regression, Least Absolute Shrinkage and Selection Operator, Elastic Net, Support Vector Regression and Gaussian Mixture Models. Four datasets were used to test the development of the SS. © 2023 IEEE.

关键词： Correlation methods

来源：评论

学校读者我要写书评

暂无评论

Strength in Diversity: Multi-Branch Representation Learning for Vehicle Re-Identification

Strength in Diversity: Multi-Branch Representation Learning ...

引用

International Conference on Intelligent Transportation

作者： Eurico Almeida Bruno Silva Jorge Batista Institute of Systems and Robotics University of Coimbra Coimbra Portugal Department of Electrical and Computers Engineering Faculty of Science and Technology University of Coimbra Coimbra Portugal

This paper presents an efficient and lightweight multi-branch deep architecture to improve vehicle reidentification (V-ReID). While most V-ReID work uses a combination of complex multi-branch architectures to extract robust and diversified embeddings towards re-identification, we advocate that simple and lightweight architectures can be designed to fulfill the Re-Idtask without compromising performance. We propose a combination of Grouped-convolution and Loss-Branch-Split strategies to design a multi-branch architecture that improve feature diversity and feature discriminability. We combine a ResNet50 global branch architecture with a BotNet self-attention branch architecture, both designed within a Loss-Branch-Split (LBS) strategy. We argue that specialized loss-branch-splitting helps to improve re-identification tasks by generating specialized re-identification features. A lightweight solution using grouped convolution is also proposed to mimic the learning of loss-splitting into multiple embeddings while significantly reducing the model size. In addition, we designed an improved solution to leverage additional metadata, such as camera ID and pose information, that uses 97% less parameters, further improving re-identification performance. In comparison to state-of-the-art (SoTA) methods, our approach outperforms competing solutions in Veri-776 by achieving 85.6% mAP and 97.7% CMC1 and obtains competitive results in Veri-Wild with 88.1% mAP and 96.3% CMC1. Overall, our work provides important insights into improving vehicle re-identification and presents a strong basis for other retrieval tasks. Our code is available at the link.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Two-Stream Architecture with Contrastive and Self-Supervised Attention Feature Fusion for Error-related Potentials Classification

Two-Stream Architecture with Contrastive and Self-Supervised...

引用

IEEE International Workshop on Robot and Human Communication (ROMAN)

作者： Luís Garrote João Perdiz Mine Yasemin Gabriel Pires Urbano J. Nunes Department of Electrical and Computer Engineering Institute of Systems and Robotics University of Coimbra Portugal Polytechnic Institute of Tomar Portugal

ISBN: (数字)9798350375022

ISBN: (纸本)9798350375039

Error-related potentials (ErrPs) extracted from electroencephalographic signals hold potential for application in Brain-Machine Interfaces, in contexts such as robot teleoperation or shared control in assistive platforms. Due to difficulties in signal classification, in part caused by its non-stationary and noisy nature, their use has not been fully realized *** work proposes a new approach to ErrP classification based on a two-stream deep learning architecture with three training stages. Its first stage is a self-supervised autoencoder architecture with a multi-head attention layer providing relevant latent features. The second stage comprises a supervised contrastive learning approach considering two backbone networks, where one inherits weights from the first stage and the other is updated by considering the feature embeddings distribution. The final stage comprises supervised classification, where the two backbones are fused and used to classify the input EEG signal. At the end of the three stages, a data-driven two-stream ErrP model is ***-five variants of the proposed approach using the Deep Convolutional Network, Shallow Convolutional Network and EEGNet backbones were tested in an ablation study and benchmarked against a large number of classical classification methods, using data from the BNCI dataset intended to assess cross subject generalization capabilities. The proposed approach obtained the best results overall, highlighting the approach’s capabilities in capturing relevant representations of the EEG signal.

关键词： Training Electric potential Deep architecture Contrastive learning Feature extraction Brain modeling Electroencephalography Noise measurement Convolutional neural networks Robots

来源：评论

学校读者我要写书评

暂无评论

Exploiting Object-based and Segmentation-based Semantic Features for Deep Learning-based Indoor Scene Classification

arXiv

引用

arXiv 2024年

作者： Pereira, Ricardo Garrote, Luís Barros, Tiago Lopes, Ana Nunes, Urbano J. University of Coimbra Institute of Systems and Robotics Department of Electrical and Computer Engineering Portugal Polytechnic Institute of Tomar Portugal

Indoor scenes are usually characterized by scattered objects and their relationships, which turns the indoor scene classification task into a challenging computer vision task. Despite the significant performance boost in classification tasks achieved in recent years, provided by the use of deep-learning-based methods, limitations such as inter-category ambiguity and intracategory variation have been holding back their performance. To overcome such issues, gathering semantic information has been shown to be a promising source of information towards a more complete and discriminative feature representation of indoor scenes. Therefore, the work described in this paper uses both semantic information, obtained from object detection, and semantic segmentation techniques. While object detection techniques provide the 2D location of objects allowing to obtain spatial distributions between objects, semantic segmentation techniques provide pixel-level information that allows to obtain, at a pixel-level, a spatial distribution and shape-related features of the segmentation categories. Hence, a novel approach that uses a semantic segmentation mask to provide Hu-momentsbased segmentation categories' shape characterization, designated by Segmentation-based Hu-Moments Features (SHMFs), is proposed. Moreover, a three-main-branch network, designated by GOS2F2App, that exploits deep-learning-based global features, object-based features, and semantic segmentation-based features is also proposed. GOS2F2App was evaluated in two indoor scene benchmark datasets: SUN RGB-D and NYU Depth V2, where, to the best of our knowledge, state-of-the-art results were achieved on both datasets, which present evidences of the effectiveness of the proposed approach. © 2024, CC BY.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Intelligent upper-limb exoskeleton integrated with soft bioelectronics and deep learning for intention-driven augmentation

引用

npj Flexible Electronics 2024年第1期8卷 856-868页

作者： Jinwoo Lee Kangkyu Kwon Ira Soltis Jared Matthews Yoon Jae Lee Hojoong Kim Lissette Romero Nathan Zavanelli Youngjin Kwon Shinjae Kwon Jimin Lee Yewon Na Sung Hoon Lee Ki Jun Yu Minoru Shinohara Frank L.Hammond Woon-Hong Yeo Department of Mechanical Roboticsand Energy EngineeringDongguk UniversitySeoul 04620Republic of Korea IEN Center for Wearable Intelligent Systems and Healthcare Institute for Electronics and NanotechnologyGeorgia Institute of TechnologyAtlantaGA 30332USA School of Electrical and Computer Engineering Georgia Institute of TechnologyAtlantaGA 30332USA George W.Woodruff School of Mechanical Engineering Georgia Institute of TechnologyAtlantaGA 30332USA School of Industrial Design Georgia Institute of TechnologyAtlantaGA 30332USA School of Electrical and Electronic Engineering Yonsei UniversitySeoul 03722Republic of Korea School of Biological Sciences Georgia Institute of TechnologyAtlantaGA 30332USA Wallace H.Coulter Department of Biomedical Engineering Georgia Institute of Technology and Emory University School of MedicineAtlantaGA 30332USA Institute for Materials Parker H.Petit Institute for Bioengineering and BiosciencesInstitute for Robotics and Intelligent MachinesGeorgia Institute of TechnologyAtlantaGA 30332USA

The age and stroke-associated decline in musculoskeletal strength degrades the ability to perform daily human tasks using the upper ***,we introduce an intelligent upper-limb exoskeleton system that utilizes deep learning to predict human intention for strength *** embedded soft wearable sensors provide sensory feedback by collecting real-time muscle activities,which are simultaneously computed to determine the user’s intended *** deep learning predicts four upper-limb joint motions with an average accuracy of 96.2%at a 500–550 ms response rate,suggesting that the exoskeleton operates just by human *** addition,an array of soft pneumatics assists the intended movements by providing 897 newtons of force while generating a displacement of 87mm at *** intent-driven exoskeleton can reduce human muscle activities by 3.7 times on average compared to the unassisted exoskeleton.

关键词： skeleton learning utilize

来源：评论

学校读者我要写书评

暂无评论

An Automatic Extraction of Features from Water Bodies by using Contrast Vision Transformer 2

An Automatic Extraction of Features from Water Bodies by usi...

引用

2nd IEEE International Conference on Integrated Intelligence and Communication systems, ICIICS 2024

作者： Madhumitha, A. Al-Fatlawi, Muhammed Sai Srinivas, T. Aditya Ramachandra, A.C. Venkatasubramanian, R. Department of Computer Science and Business Systems Bannari Amman Institute of Technology Sathyamangalam Erode India Department of Computers Techniques Engineering College of Technical Engineering The Islamic University Najaf Iraq Department of Artificial Intelligence and Machine Learning Jayaprakash Naryan College of Engineering Mahabubnagar India Department of Electronics and Communication Engineering Nitte Meenakshi Institute of Technology Bengaluru India Department of Electrical and Electronics Engineering New Prince Shri Bhavani College of Engineering and Technology Chennai India

ISBN: (纸本)9798331504960

In recent days, for strengthening the climate change adaption water features are utilizing as the most crucial environmental element. The integration of Remote Sensing (RS) with Artificial Intelligence (AI) is becoming the major sought-after approach for extracting the features of water information. In existing, the Faster Regional Convolution Neural Network (FR_CNN) is utilized for the water bodies’ extraction. However, the existing FR_CNN model struggled with computational complexity. To overcome this issue, Contrast Vision Transformer (CoVT) is proposed to extract features from water bodies. The transformer with self-attention mechanism reduces the computational complexity by eliminating regional proposal networks. Initially, the satellite images are taken from sentinel-2 dataset and further processed into preprocessing. In preprocessing Gaussian Median Filter (GMF) and Atmospheric Correction (AC) are utilized to reduce the noise and atmospheric effects from images. After that, proposed CoVT extract features from preprocessed images and fed into projection head to generate projected features. Finally, the projected features are further processed with prediction head to predict the satellite image features of water bodies. From the results, the proposed CoVT provided better results when compared to existing FR_CNN by providing results such as accuracy (99.4%), precision (99.1%), recall (98.5%) and F1-score (98.7%) respectively. © 2024 IEEE.

关键词： Median filters

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：