检索结果-内蒙古大学图书馆

23rd International Conference on Control, Automation and Systems, ICCAS 2023

作者： Ha, Jin-Ung Kim, Hyun-Woo Cho, Myungjin Lee, Min-Chul Kyushu Institute of Technology Department of Computer Science and Networks Fukuoka Iizuka820-8502 Japan School of ICT Robotics and Mechanical Engineering Iitc Hankyong National University Kyonggi-do Anseong17579 Korea Republic of

ISBN: (纸本)9788993215274

In this paper, we propose a method to improve the recognition rate of ArUco markers under photon-starved conditions by using preprocessing such as photon counting algorithm and histogram equalization to estimate the exact three-dimensional(3D) location of improved objects. In the conventional ArUco marker recognition technology, the recognition rate of ArUco markers decreases under photon-starved conditions, and the exact 3D location of the object cannot be estimated. In the conventional ArUco marker recognition technology, the photon counting algorithm can be used to estimate the 3D location of objects by recognizing ArUco markers under photon-starved conditions. However, the photon counting algorithm has the problem that the computation speed is slowed down when there are many photons to estimate, and the recognition rate of the ArUco marker is reduced when there are few photons to estimate. To solve these problems of the photon counting algorithm and improve the recognition rate of ArUco markers, we apply histogram equalization to ArUco markers with the photon counting algorithm. To verify the feasibility of the proposed method, we implement simulations using the Open Manipulator-X platform and show performance metrics such as Peak Signal-to-Noise Ratio (PSNR). © 2023 ICROS.

关键词： Photons

来源：评论

学校读者我要写书评

暂无评论

PARE-Net: Position-Aware Rotation-Equivariant Networks for Robust Point Cloud Registration

arXiv

引用

arXiv 2024年

作者： Yao, Runzhao Du, Shaoyi Cui, Wenting Tang, Canhui Yang, Chengwu National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University China

Learning rotation-invariant distinctive features is a fundamental requirement for point cloud registration. Existing methods often use rotation-sensitive networks to extract features, while employing rotation augmentation to learn an approximate invariant mapping rudely. This makes networks fragile to rotations, overweight, and hinders the distinctiveness of features. To tackle these problems, we propose a novel position-aware rotation-equivariant network, for efficient, light-weighted, and robust registration. The network can provide a strong model inductive bias to learn rotation-equivariant/invariant features, thus addressing the aforementioned limitations. To further improve the distinctiveness of descriptors, we propose a position-aware convolution, which can better learn spatial information of local structures. Moreover, we also propose a feature-based hypothesis proposer. It leverages rotation-equivariant features that encode fine-grained structure orientations to generate reliable model hypotheses. Each correspondence can generate a hypothesis, thus it is more efficient than classic estimators that require multiple reliable correspondences. Accordingly, a contrastive rotation loss is presented to enhance the robustness of rotation-equivariant features against data degradation. Extensive experiments on indoor and outdoor datasets demonstrate that our method significantly outperforms the SOTA methods in terms of registration recall while being lightweight and keeping a fast speed. Moreover, experiments on rotated datasets demonstrate its robustness against rotation variations. Code is available at https://***/yaorz97/PARENet. Copyright © 2024, The Authors. All rights reserved.

关键词： HTTP

来源：评论

学校读者我要写书评

暂无评论

Accelerated learning and co-optimization of elastocaloric effect and stress hysteresis of elastocaloric alloys

引用

Rare Metals 2024年第12期43卷 6606-6624页

作者： Shi-Yu He Fei Xiao Rui-Hang Hou Shun-Gui Zuo Ying Zhou Xiao-Rong Cai Zhu Li Yan-Ming Wang Aysu Catal-Isik Enrique Galindo-Nava Xue-Jun Jin State Key Lab of Metal Matrix Composite School of Materials Science and EngineeringShanghai Jiao Tong UniversityShanghai200240China Materials Genome Initiative Center Shanghai Jiao Tong UniversityShanghai200240China Institute of Medical Robotics Shanghai Jiao Tong UniversityShanghai200240China Department of Materials Science and Engineering Graduate School of EngineeringOsaka UniversityOsaka565-0871Japan University of Michigan-Shanghai Jiao Tong University Joint Institute Shanghai Jiao Tong UniversityShanghai200240China Department of Mechanical Engineering University College LondonTorrington PlaceLondonWC1E 7JEUK

The development of high-performance advanced elastocaloric alloys is crucial for the implementation of elastocaloric ***,we present a machine learning(ML)framework to accelerate the development of novel elastocaloric alloys with large adiabatic temperature change(ΔT_(ad))and low stress hysteresis(Δσ_(hy)).The comprehensive framework comprises database construction,feature selection,model construction,alloy design and validation,and model *** are selected according to the physical attributes they *** that may reflect the compatibility between parent and product phases,lattice distortion and the free energy in the alloy are considered in the *** them,the key features are selected by recursive feature elimination and exhaustive search *** trained models in combination with the Bayesian optimization method are exploited to achieve multi-objective *** to the results,a newly designed elastocaloric alloy shows a large adiabatic temperature change of 15.2 K and low average stress hysteresis of70.3 MPa at room temperature,which is consistent with our *** predictions of our ML model are interpreted by the Shapley additive exPlainations(SHAP)approach,which explicitly quantifies the effects of each feature in our model on the adiabatic temperature change and stress ***,we employ the Sureindependence screening and sparsifying operator(SISSO)method in conjunction with the key features to formulate explicit model.

关键词： Elastocaloric alloys Adiabatic temperature change Stress hysteresis Machine learning Multi-objective optimization

来源：评论

学校读者我要写书评

暂无评论

Current Status,Challenges,and Prospects for New Types of Aerial Robots

引用

engineering 2024年第10期41卷 19-34页

作者： Xidong Zhou Hang Zhong Hui Zhang Wei He Hean Hua Yaonan Wang College of Robotics Hunan UniversityChangsha 410082China National Engineering Research Center of Robot Visual Perception and Control Technology Hunan UniversityChangsha 410082China School of Intelligence Science and Technology Beijing University of Science and TechnologyBeijing 100083China

New types of aerial robots(NTARs)have found extensive applications in the military,civilian contexts,scientific research,disaster management,and various other *** with traditional aerial robots,NTARs exhibit a broader range of morphological diversity,locomotion capabilities,and enhanced operational ***,this study defines aerial robots with the four characteristics of morphability,biomimicry,multi-modal locomotion,and manipulator attachment as ***,this paper discusses the latest research progress in the materials and manufacturing technology,actuation technology,and perception and control technology of ***,the research status of NTAR systems is summarized,focusing on the frontier development and application cases of flapping-wing microair vehicles,perching aerial robots,amphibious robots,and operational aerial ***,the main challenges presented by NTARs in terms of energy,materials,and perception are analyzed,and the future development trends of NTARs are summarized in terms of size and endurance,mechatronics,and complex scenarios,providing a reference direction for the follow-up exploration of NTARs.

关键词： Aerial robot Morphability Biomimicry Perch Amphibious

来源：评论

学校读者我要写书评

暂无评论

MLS: A Meta-Learning Based Stackelberg Model for Robot Trajectory Guidance

MLS: A Meta-Learning Based Stackelberg Model for Robot Traje...

引用

robotics, Control and Automation (ICRCA), International Conference on

作者： Jin Guo Zhiyong Jiang Yifan Liu Lili Gong Robotics Engineering Center The 21st Research Institute of China Electronics Technology Group Corporation Shanghai China

ISBN: (数字)9798350344721

ISBN: (纸本)9798350344738

The critical role of trajectory guidance in facilitating follower robots to reach target destinations under the supervision of a leader robot is well-established. However, the complexity of coordinating efficient collaboration increases when a leader must guide a diverse set of followers with unknown specific characteristics. To address this challenge, we propose a novel meta-learning-based Stackelberg approach. We model the guided trajectory planning problem as a dynamic Stackelberg game, capturing the intricate interaction dynamics between the leader and followers. Utilizing meta-learning techniques, the leader learns a meta-best-response model from a selected set of followers. When a follower initiates a guidance request, the leader swiftly adapts to the follower's unique characteristics using a few learning data. This follower-specific model is then used for effective trajectory guidance. Comprehensive simulations demonstrate the superior generalization and adaptability of our proposed method over alternative learning methods, and emphasize the practical value and effectiveness of the guidance mechanism. Our findings underscore the potential of our meta-learning-based Stackelberg approach in addressing collaborative trajectory planning challenges in heterogeneous multi-robot systems.

关键词： Metalearning Adaptation models Trajectory planning Robot kinematics Collaboration Games Trajectory

来源：评论

学校读者我要写书评

暂无评论

FinFix: A Soft Gripper With Contact-Reactive Reflex for High-Speed Pick and Place of Fragile Objects

FinFix: A Soft Gripper With Contact-Reactive Reflex for High...

引用

IEEE International Conference on Soft robotics (RoboSoft)

作者： Willem Heeringa Cosimo Della Santina Gerwin Smit BioMechanical Engineering department Delft University of Technology Delft Netherlands Cognitive Robotics department Delft University of Technology Institute of Robotics and Mechatronics Delft Netherlands German Aerospace Center (DLR) Institute of Robotics and Mechatronics Weßling Germany

Industrial automation calls for precise tasks with cycle times reduced to the minimum. At the same time, when handling delicate products such as fruits and vegetables, accelerations must be kept low to keep interaction forces under a certain threshold to avoid damage. This trade-off hinders the penetration of automation in many relevant application fields. This paper investigates using soft technology to solve this challenge. We propose the FinFix gripper, a non-anthropomorphic soft gripper capable of handling delicate objects at high acceleration using a contact-reactive grasping approach. This gripper has two entirely passive sensorized fingers that establish contact and two active fingers that are actuated pneumatically through a rigid mechanism allowing for rapid closure. We provide exhaustive experimental validation by connecting the gripper to a delta robot. The system can reliably execute pick-and-place cycles in $\sim 1 \ \mathrm{s}$ when the distance between the pick and the place locations is 400 mm, resulting in a peak speed of $\sim 10\frac{\mathrm{m}}{\mathrm{s}}$ . None of the fragile objects used during the experiments showed any damage. The only information needed is a rough estimation of the object's position to be grasped and a contact event to trigger the reflex. The test results show that the gripper can hold fragile objects during lateral accelerations of 10g.

关键词： Automation Grasping Soft robotics Robot sensing systems Sensors Reliability Grippers

来源：评论

学校读者我要写书评

暂无评论

FEIF: Feature Excitation and Interactive Fusion for 6D Object Pose Estimation

FEIF: Feature Excitation and Interactive Fusion for 6D Objec...

引用

2022 IEEE International Conference on robotics and Biomimetics, ROBIO 2022

作者： Meng, Qiwei Zhu, Shiqiang Gu, Jianjun Lin, Zheyuan Zhong, Ling Song, Wei Research Center for Intelligent Robotics Research Institute of Interdisciplinary Innovation Zhejiang Laboratory Hangzhou311100 China Zhejiang Engineering Research Center for Intelligent Robotics Hangzhou311100 China Dalhousie University Department of Electrical and Computer Engineering HalifaxNSB3H 4R2 Canada

ISBN: (纸本)9781665481090

In this paper, we present FEIF, a novel framework for 6D object pose estimation from a single RGBD image. Compared with previous approaches, this model fully leverages the complementary RGB and depth information through lay-erwise feature excitation and interactive fusion, thus improving the effectiveness and robustness of extracted features. The representative RGBD features can be subsequently applied for object keypoints prediction and pose computation. In addition, our FEIF includes a self-evaluation module to make robots aware of the confidence of pose estimation, so they are flexible to take the corresponding strategy for following tasks. Experiment results prove the SOTA performance of our model on LineMOD, Occlusion-LineMOD and YCB-Video datasets, and we also demonstrate its applicability and robustness in real-world robotic manipulation. © 2022 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Augmenting Transformer Autoencoders with Phenotype Classification for Robust Detection of Psychotic Relapses

Augmenting Transformer Autoencoders with Phenotype Classific...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： N. Efthymiou G. Retsinas P. P. Filntisis P. Maragos Institute of Robotics Athena Research and Innovation Center Maroussi Greece School of ECE National Technical University of Athens Athens Greece

Recently, deep autoencoder architectures have received attention for the problem of unsupervised anomaly detection. Detecting psychotic relapses in mental health patients is a crucial challenge, often framed as anomaly detection, given the limited availability of data during relapsing states. In this paper, motivated by the fact that during relapses patients tend to undergo behavioral changes, we augment the classical autoencoder architecture with extra patient identification components. We show that formulating the problem as one of both signal reconstruction and patient identification largely improves the overall precision and robustness of relapse detection and significantly outperforms previous methods with a relative improvement of 15%. In addition, we also explore multiple ways to fuse the identification and reconstruction errors into a unified anomaly score that outperforms the results achieved by each error in isolation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Accurate modulation of photoprinting under stiffness imaging feedback for engineering ECMs with high-fidelity mechanical properties

引用

Microsystems & Nanoengineering 2022年第3期8卷 291-301页

作者： Xin Li Huaping Wang Xinyi Dong Qing Shi Tao Sun Shingo Shimoda Qiang Huang Toshio Fukuda Intelligent Robotics Institute School of Mechatronical EngineeringBeijing Institute of TechnologyBeijing 100081China The Key Laboratory of Biomimetic Robots and Systems(Beijing Institute of Technology) Ministry of EducationBeijing 100081China School of Medical Technology Beijing Institute of TechnologyBeijing 100081China Intelligent Behavior Control Collaboration Unit RIKEN Center of Brain Science463-0003 NagoyaJapan

Engineered extracellular matrices(ECMs)that replicate complex in-vivo features have shown great potential in tissue *** hydrogel microstructures have been widely used to replace these native ECMs for physiologically relevant ***,accurate reproduction of the 3D hierarchical and nonuniform mechanical stffness inside one integrated microstructure to mimic the complex mechanical properties of native ECMs presents a major ***,by using digital holographic microscopy(DHM)-based stffness imaging feedback,we propose a novel closed-loop control algorithm to achieve high-accuracy control of mechanical properties for hydrogel microstructures that recapitulate the physiological properties of native ECMs with high *** photoprinting,the photocuring area of the hydrogel is divided into microscale grid areas to locally control the photocuring *** the assistance of a motorized microfluidic channel,the curing thickness is controlled with layer-by-layer *** DHM-based stiffness imaging feedback allows accurate adjustment of the photocuring degree in every grid area to change the crosslinking network density of the hydrogel,thus enabling large-span and high-resolution modulation of mechanical ***,the gelatin methacrylate was used as a typical biomaterial to construct the highfidelity biomimetic *** Young's modulus could be flexibly modulated in the 10 kPa to 50 kPa ***,the modulus gradient was accurately controlled to within 2.9 *** engineering ECM with locally different mechanical properties,cell spreading along the stff areas was observed *** believe that this method can regenerate complex biomimetic ECMs that closely recapitulate in-vivo mechanical properties for further applications in tissue engineering and biomedical research.

关键词： microstructure ECM curing

来源：评论

学校读者我要写书评

暂无评论

Nearly Lossless Adaptive Bit Switching

arXiv

引用

arXiv 2025年

作者： Huang, Haiduo Liu, Zhenhua Xia, Tian Zhao, Wenzhe Ren, Pengju The National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center of Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Shaanxi Xi’an China

Model quantization is widely applied for compressing and accelerating deep neural networks (DNNs). However, conventional Quantization-Aware Training (QAT) focuses on training DNNs with uniform bit-width. The bit-width settings vary across different hardware and transmission demands, which induces considerable training and storage costs. Hence, the scheme of one-shot joint training multiple precisions is proposed to address this issue. Previous works either store a larger FP32 model to switch between different precision models for higher accuracy or store a smaller INT8 model but compromise accuracy due to using shared quantization parameters. In this paper, we introduce the Double Rounding quantization method, which fully utilizes the quantized representation range to accomplish nearly lossless bit-switching while reducing storage by using the highest integer precision instead of full precision. Furthermore, we observe a competitive interference among different precisions during one-shot joint training, primarily due to inconsistent gradients of quantization scales during backward propagation. To tackle this problem, we propose an Adaptive Learning Rate Scaling (ALRS) technique that dynamically adapts learning rates for various precisions to optimize the training process. Additionally, we extend our Double Rounding to one-shot mixed precision training and develop a Hessian-Aware Stochastic Bit-switching (HASB) strategy. Experimental results on the ImageNet-1K classification demonstrate that our methods have enough advantages to state-of-the-art one-shot joint QAT in both multi-precision and mixed-precision. We also validate the feasibility of our method on detection and segmentation tasks, as well as on LLMs task. Our codes are available at https://***/haiduo/Double-Rounding. Copyright © 2025, The Authors. All rights reserved.

关键词： Deep neural networks

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：