检索结果-内蒙古大学图书馆

27th International Conference on System Theory, Control and Computing, ICSTCC 2023

作者： Yadav, Harsh Xue, Honghu Rudall, Yan Bakr, Mohamed Hein, Benedikt Rueckert, Elmar Nguyen, Ngoc Thinh Inst. for Robotics and Cognitive Systems University of Lübeck Luebeck Germany Robotics and Automation KION Group AG Hamburg Germany Inst. of Automation Technology Helmut-Schmidt-University Hamburg Germany Inst. for Cyber Physical Systems Montanuniversität Leoben Leoben Austria

ISBN: (纸本)9798350337983

This paper presents a study on the mapless navigation of autonomous mobile robot using Deep Reinforcement Learning in an intralogistics setting. The task is to make an autonomous mobile robot learn to navigate to a goal without prior knowledge of the environment. In this paper, a controller using the Soft Actor-Critic algorithm is designed, trained, and applied for navigating the robot equipped with 360° LiDAR and front camera sensors. The controller is successfully validated in an almost fully observable environment under extensive simulations. Furthermore, we investigate the performance of the proposed controller in a partially observable environment and possible limitations. We use a 3D Temporal Convolution Network for processing the time series image data from visual observations. Besides Partial Observability, we also address the problem of sparse positive rewards in training the Deep Reinforcement Learning algorithm with a combined approach of Automatic Curriculum Learning and Dual Prioritized Experience Replay. © 2023 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A Data-Centric Approach for Object Recognition in Hemispherical Camera Images 49

A Data-Centric Approach for Object Recognition in Hemispheri...

引用

49th Annual Conference of the IEEE Industrial Electronics Society, IECON 2023

作者： Balgabekova, Zarema Alaran, Muslim Varol, Huseyin Atakan Inst. of Smart Systems and Ai Nazarbayev University Astana Kazakhstan Nazarbayev University Dept. of Robotics and Mechatronics Astana Kazakhstan

ISBN: (纸本)9798350331820

Object recognition is a machine learning problem that involves the correct classification and localization of objects in an image. Object recognition has found wide applications in Industry 4.0, surveillance, and autonomous driving. Prominent datasets such as Pascal VOC, ImageNet, and MS COCO have further spurred advances in object recognition. However, the images in these datasets were captured using ordinary perspective lenses. Unlike ordinary cameras, fisheye or hemispherical cameras have a wide FoV reaching 180°. Due to this property, they can acquire wide panoramic images, thus, capturing more objects in a scene as compared to ordinary perspective lenses. In this work, we provide the first benchmark testing dataset of real fisheye images containing commonly-occurring objects in the human environment obtained with hemispherical cameras. The dataset comprises 1,000 images containing 39 classes with a total number of 14,218 object inst.nces. We trained the YOLOv7 model on the original MS COCO and transformed FisheyeCOCO datasets and tested the obtained models with our dataset. The results indicate that training the model on the combination of original and transformed datasets improves the performance by 2.5% compared to using these datasets separately. © 2023 IEEE.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

Automated Ultrasonic Testing of Carbon-Carbon Composite Materials 7

Automated Ultrasonic Testing of Carbon-Carbon Composite Mate...

引用

7th International Conference on Information, Control, and Communication Technologies, ICCT 2023

作者： Pavlukhin, Evgeniy A. Kinzhagulov, Igor Y. Fedorov, Alexey V. Omarov, Magomed A. Itmo University Faculty of Control Systems and Robotics ST Petersburg Russia Inst. of Science 'Engineering and Design Center for Support of the Operation of Space Technology ST Petersburg Russia

ISBN: (纸本)9798350340945

The quality of products made of carbon-carbon composite materials (CCCM) is influenced by many factors, including the parameters of the technological process and the technical level of equipment for the manufacture of materials and structures. One of the most important moments in the process of creating products from CCCM is the formation of reinforcing components, matrix and their structural bonds. The efficiency of the system of structural bonds depends on the homogeneity (heterogeneity) of the material structure. The required mechanical and thermal properties of CCCM are ensured by the uniformity (continuity) of the structure, the change of which affects the properties of the material as a whole. In the event of discontinuity, an uneven distribution of mechanical and thermal fields can occur, which will jeopardize the performance of the entire structure, so non-destructive testing (NDT) of CCCM is an important task. The ultrasonic shadow method of control was chosen as the NDT method. This method makes it possible to detect structural inhomogeneities in the material of the test object - accumulations of pores and microcracks (at the microlevel), cracks, zones of low density, local heterogeneity of the structure (at the macrolevel). This article discusses the automation and ultrasonic testing of products from CCCM. © 2023 IEEE.

关键词： Nondestructive examination

来源：评论

学校读者我要写书评

暂无评论

Probing Local Cellular Mechanics by Atomic Force Microscopy with Modified Spherical Tip

Probing Local Cellular Mechanics by Atomic Force Microscopy ...

引用

2022 IEEE International Conference on Cyborg and Bionic systems, CBS 2022

作者： Zuo, Zhaofeng Liu, Xiaoming Tang, Xiaoqing Liu, Fengyu Liu, Dan Li, Yuyang Huang, Qiang Arai, Tatsuo Beijing Institute of Technology Beijing Adv. Innov. Ctr. for Intelligent Robots and Systems The Intelligent Robotics Inst. School of Mechatronical Engineering Beijing100081 China

ISBN: (纸本)9781665490283

The mechanical characterization of cell is important for knowing the physiological state and studying diseases of organism. While many approaches are available for measuring cellular elasticity, distinguishing the stiffness variation among different cellular areas is still a challenge. In this paper, we reported a method to modify spherical atomic force microscopy (AFM) tip for accurate measurement of Young's modulus in several areas on single adherent living cancer cells (Hela cells). The micrometer size spheres were transported to an AFM probe tip by dual micropipettes and fixed by ultraviolet (UV) curable glue, which is reproducible and nondestructive to the cantilever. The force-displacement curves were measured along major axis of cells with the modified AFM probe in indentation experiments. The results demonstrate that the modulus value increased with the detected point approaching the nucleus center, and areas closed to nucleus showed have higher stiffness. Our study provides a quantitative method for static measurement of different locations of cell, and the result has the potential to reveal how intracellular structures have effect on the cell mechanical characterization. © 2023 IEEE.

关键词： Probes

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning for Mapless Navigation of Autonomous Mobile Robot

Deep Reinforcement Learning for Mapless Navigation of Autono...

引用

International Conference on System Theory, Control, and Computing (ICSTCC)

作者： Harsh Yadav Honghu Xue Yan Rudall Mohamed Bakr Benedikt Hein Elmar Rueckert Ngoc Thinh Nguyen Inst. for Robotics and Cognitive Systems University of Lübeck Luebeck Germany Robotics and Automation KION Group AG Hamburg Germany Inst. of Automation Technology Helmut-Schmidt-University Hamburg Germany Inst. for Cyber Physical Systems Montanuniversität Leoben Leoben Austria

This paper presents a study on the mapless navigation of autonomous mobile robot using Deep Reinforcement Learning in an intralogistics setting. The task is to make an autonomous mobile robot learn to navigate to a goal without prior knowledge of the environment. In this paper, a controller using the Soft Actor-Critic algorithm is designed, trained, and applied for navigating the robot equipped with $360^{\mathrm{o}}$ LiDAR and front camera sensors. The controller is successfully validated in an almost fully observable environment under extensive simulations. Furthermore, we investigate the performance of the proposed controller in a partially observable environment and possible limitations. We use a 3D Temporal Convolution Network for processing the time series image data from visual observations. Besides Partial Observability, we also address the problem of sparse positive rewards in training the Deep Reinforcement Learning algorithm with a combined approach of Automatic Curriculum Learning and Dual Prioritized Experience Replay.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Data-Centric Approach for Object Recognition in Hemispherical Camera Images

A Data-Centric Approach for Object Recognition in Hemispheri...

引用

Annual Conference of Industrial Electronics Society

作者： Zarema Balgabekova Muslim Alaran Huseyin Atakan Varol Inst. of Smart Systems and AI Nazarbayev University Astana Kazakhstan Dept. of Robotics and Mechatronics Nazarbayev University Astana Kazakhstan

关键词：

来源：评论

学校读者我要写书评

暂无评论

AnyFace: A Data-Centric Approach For Input-Agnostic Face Detection

AnyFace: A Data-Centric Approach For Input-Agnostic Face Det...

引用

International Conference on Big Data and Smart Computing (BIGCOMP)

作者： Askat Kuzdeuov Darina Koishigarina Huseyin Atakan Varol Inst. of Smart Systems and AI Nazarbayev University Astana Kazakhstan Dept. of Robotics and Mechatronics Nazarbayev University Astana Kazakhstan

Face detection is a mandatory step in many computer vision applications, such as face recognition, emotion recognition, age detection, virtual makeup, and vital sign monitoring. Thanks to advancements in deep learning and the introduction of annotated large-scale datasets, numerous applications have been developed for human faces. Recently, other domains, such as animals and cartoon characters, have started gaining attention but still lag far behind human faces. The biggest challenge is the limited number of annotated face datasets in these domains. The manual labeling of large-scale datasets is tedious and requires substantial human labor. In this regard, we present an inputagnostic face detector to ease the annotation of various face datasets. We propose a simple but effective data-centric approach inst.ad of building a specific neural network architecture. Specifically, we trained a face detection model, YOLO5Face, on human, animal, and cartoon face datasets. The experiments show that the model can achieve accurate results in all domains. In addition, the model achieved decent results for animals and cartoon characters different from the ones in the training set. This implies that the model can extract agnostic facial features. We have made the source code and pre-trained models publicly available at https://***/IS2AI/AnyFace to stimulate research in these fields.

关键词： Training Animals Face recognition Source coding Neural networks Detectors Data models

来源：评论

学校读者我要写书评

暂无评论

A Data-Centric Approach for Object Recognition in Hemispherical Camera Images

TechRxiv

引用

TechRxiv 2023年

作者： Balgabekova, Zarema Alaran, Muslim Varol, Huseyin Atakan Inst. of Smart Systems and AI Nazarbayev University Astana Kazakhstan Dept. of Robotics and Mechatronics Nazarbayev University Astana Kazakhstan

Object recognition is a machine learning problem that involves the correct classification and localization of objects in an image. Object recognition has found wide applications in Industry 4.0, surveillance, and autonomous driving. Prominent datasets such as Pascal VOC, ImageNet, and MS COCO have further spurred advances in object recognition. However, the images in these datasets were captured using ordinary perspective lenses. Unlike ordinary cameras, fisheye or hemispherical cameras have a wide FoV reaching 180◦. Due to this property, they can acquire wide panoramic images, thus, capturing more objects in a scene as compared to ordinary perspective lenses. In this work, we provide the first benchmark testing dataset of real fisheye images containing commonly-occurring objects in the human environment obtained with hemispherical cameras. The dataset comprises 1,000 images containing 39 classes with a total number of 14,218 object inst.nces. We trained the YOLOv7 model on the original MS COCO and transformed FisheyeCOCO datasets and tested the obtained models with our dataset. The results indicate that training the model on the combination of original and transformed datasets improves the performance by 2.5% compared to using these datasets separately. © 2023, CC BY.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

AnyFace: A Data-Centric Approach For Input-Agnostic Face Detection

TechRxiv

引用

TechRxiv 2022年

作者： Kuzdeuov, Askat Koishigarina, Darina Varol, Huseyin Atakan Inst. of Smart Systems and Artificial Intelligence Nazarbayev University Astana Kazakhstan Dept. of Robotics and Mechatronics Nazarbayev University Astana Kazakhstan

Face detection is a mandatory step in many computer vision applications, such as face recognition, emotion recognition, age detection, virtual makeup, and vital sign monitoring. Thanks to advancements in deep learning and the introduction of annotated large-scale datasets, numerous applications have been developed for human faces. Recently, other domains, such as animals and cartoon characters, have started gaining attention but still lag far behind human faces. The biggest challenge is the limited number of annotated face datasets in these domains. The manual labeling of large-scale datasets is tedious and requires substantial human labor. In this regard, we present an input-agnostic face detector to ease the annotation of various face datasets. We propose a simple but effective data-centric approach inst.ad of building a specific neural network architecture. Specifically, we trained a face detection model, YOLO5Face, on human, animal, and cartoon face datasets. The experiments show that the model can achieve accurate results in all domains. In addition, the model achieved decent results for animals and cartoon characters different from the ones in the training set. This implies that the model can extract agnostic facial features. We have made the source code and pre-trained models publicly available at https://***/IS2AI/AnyFace to stimulate research in these fields. © 2022-12-01, CC BY.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Sparse Gaussian Process Regression using Progressively Growing Learning Representations

Sparse Gaussian Process Regression using Progressively Growi...

引用

IEEE Conference on Decision and Control

作者： Christos N. Mavridis George P. Kontoudis John S. Baras Department of Electrical and Computer Engineering Institute for Systems Research University of Maryland College Park USA Dept. of Aerospace Engineering Maryland Robotics Center Inst. of Systems Research University of Maryland College Park MD USA

ISBN: (数字)9781665467612

ISBN: (纸本)9781665467629

We present a new sparse Gaussian process regression model whose covariance function is parameterized by the locations of a progressively growing set of pseudo-inputs generated by an online deterministic annealing optimization algorithm. A series of entropy-regularized optimization problems is solved sequentially, introducing a bifurcation phenomenon, according to which, pseudo-inputs are gradually generated. This results in an active learning approach, which, in contrast to most existing works, can modify already selected pseudo-inputs and is trained using a recursive gradient-free stochastic approximation algorithm. Finally, the proposed algorithm is able to incorporate prior knowledge in the form of a probability density, according to which new observations are sampled. Experimental results showcase the efficacy and potential advantages of the proposed methodology.

关键词： Annealing Gaussian processes Bifurcation Approximation algorithms Optimization

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：