检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Yang, Fan Chen, Wenrui Yang, Kailun Lin, Haoran Luo, DongSheng Tang, Conghui Li, Zhiyong Wang, Yaonan The School of Robotics Hunan University Changsha410012 China The National Engineering Research Center of Robot Visual Perception and Control Technology Hunan University Changsha410082 China

To enable robots to use tools, the initial step is teaching robots to employ dexterous gestures for touching specific areas precisely where tasks are performed. Affordance features of objects serve as a bridge in the functional interaction between agents and objects. However, leveraging these affordance cues to help robots achieve functional tool grasping remains unresolved. To address this, we propose a granularity-aware affordance feature extraction method for locating functional affordance areas and predicting dexterous coarse gestures. We study the intrinsic mechanisms of human tool use. On one hand, we use fine-grained affordance features of object-functional finger contact areas to locate functional affordance regions. On the other hand, we use highly activated coarse-grained affordance features in hand-object interaction regions to predict grasp gestures. Additionally, we introduce a model-based post-processing module that includes functional finger coordinate localization, finger-to-end coordinate transformation, and force feedback-based coarse-to-fine grasping. This forms a complete dexterous robotic functional grasping framework GAAF-Dex, which learns Granularity-Aware Affordances from human-object interaction for tool-based Functional grasping in Dexterous robotics. Unlike fully-supervised methods that require extensive data annotation, we employ a weakly supervised approach to extract relevant cues from exocentric (Exo) images of hand-object interactions to supervise feature extraction in egocentric (Ego) images. Correspondingly, we have constructed a small-scale dataset, FAH, which includes near 6K images of functional hand-object interaction Exo images and Ego images of 18 commonly used tools performing 6 tasks. Extensive experiments on the dataset demonstrate that our method outperforms state-of-the-art methods, and real-world localization and grasping experiments validate the practical applicability of our approach. The source code and the establishe

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

Learning visual-based deformable object rearrangement with local graph neural networks

arXiv

引用

arXiv 2023年

作者： Deng, Yuhong Wang, Xueqian Chen, Lipeng Tencent Robotics X Lab Shenzhen China The Center for Intelligent Control and Telescience Tsinghua Shenzhen International Graduate School Shenzhen China

Goal-conditioned rearrangement of deformable objects (e.g. straightening a rope and folding a cloth) is one of the most common deformable manipulation tasks, where the robot needs to rearrange a deformable object into a prescribed goal configuration with only visual observations. These tasks are typically confronted with two main challenges: the high dimensionality of deformable configuration space and the underlying complexity, nonlinearity and uncertainty inherent in deformable dynamics. To address these challenges, we propose a novel representation strategy that can efficiently model the deformable object states with a set of keypoints and their interactions. We further propose local-graph neural network (GNN), a light local GNN learning to jointly model the deformable rearrangement dynamics and infer the optimal manipulation actions (e.g. pick and place) by constructing and updating two dynamic graphs. Both simulated and real experiments have been conducted to demonstrate that the proposed dynamic graph representation shows superior expressiveness in modeling deformable rearrangement dynamics. Our method reaches much higher success rates on a variety of deformable rearrangement tasks (96.3% on average) than state-of-the-art method in simulation experiments. Besides, our method is much more lighter and has a 60% shorter inference time than state-of-the-art methods. We also demonstrate that our method performs well in the multi-task learning scenario and can be transferred to real-world applications with an average success rate of 95% by solely fine tuning a keypoint detector. A supplementary video can be found at https://***/AhwTQo6fCM0. Copyright © 2023, The Authors. All rights reserved.

关键词： Dynamics

来源：评论

学校读者我要写书评

暂无评论

Graph-Transporter: A Graph-based Learning Method for Goal-Conditioned Deformable Object Rearranging Task

arXiv

引用

arXiv 2023年

作者： Deng, Yuhong Xia, Chongkun Wang, Xueqian Chen, Lipeng Tencent Robotics X Lab Shenzhen China The Center for Intelligent Control and Telescience Tsinghua Shenzhen International Graduate School Shenzhen China

Rearranging deformable objects is a long-standing challenge in robotic manipulation for the high dimensionality of configuration space and the complex dynamics of deformable objects. We present a novel framework, Graph-Transporter, for goal-conditioned deformable object rearranging tasks. To tackle the challenge of complex configuration space and dynamics, we represent the configuration space of a deformable object with a graph structure and the graph features are encoded by a graph convolution network. Our framework adopts an architecture based on Fully Convolutional Network (FCN) to output pixel-wise pick-and-place actions from only visual input. Extensive experiments have been conducted to validate the effectiveness of the graph representation of deformable object configuration. The experimental results also demonstrate that our framework is effective and general in handling goal-conditioned deformable object rearranging tasks. Copyright © 2023, The Authors. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

A Novel Bio-Intelligent Real-Time Redundancy Resolution Method for Cable-Driven Parallel Manipulators

A Novel Bio-Intelligent Real-Time Redundancy Resolution Meth...

引用

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

作者： S. Ali A. Moosavian SeyedSina Seyedi HasanAbadi Arman Gholibeikian Department of Mechanical Engineering K. N. Toosi University of Technology Advanced Robotic and Automated Systems (ARAS) Center of Excellence in Robotics and Control Tehran Iran

ISBN: (数字)9798331529734

ISBN: (纸本)9798331529741

Cable-driven parallel manipulators (CDPMs) are widely used where there is a need for high speeds and high accelerations, thanks to the low inertia and low cost of their links. However, cables can only exert tensional force. Due to this inherent property of the cables, CDPMs are required to be designed with at least one degree of redundancy in their joint space, to be able to generate an acceptable range of wrench within the workspace. Meanwhile, redundancy in joint space leads a problem called redundancy resolution, which is basically a force distribution from workspace to joint space. In this research, inspired by a simple mathematical model of the limbic system of the brain, a novel real-time redundancy resolution method is proposed. The presented method utilizes the kinematic information of the end-effector to find the solution to the redundancy resolution problem. The results show that this Brain Emotional Learning (BEL) method has a better performance in comparison with some common numerical optimization methods solved with a nonlinear programming solver like fmincon function in MATLAB, in terms of computational cost, continuity of the profile of the cables forces, and robustness.

关键词： Power cables Redundancy Force Optimization methods Kinematics Brain modeling Mathematical models Real-time systems Robustness Trajectory

来源：评论

学校读者我要写书评

暂无评论

A Cost-Effective Smart Labor Assistance Trolley for Industrial Applications

A Cost-Effective Smart Labor Assistance Trolley for Industri...

引用

International Conference on robotics and Artificial Intelligence (ICRAI)

作者： Muneeb Zafar Sarmad Shafique Faisal Riaz Samia Abid Umar Raza William Holderbaum Department of CS & IT National Center of Robotics and Automation (NCRA) Control Automotive and Robotics (CAR) Mirpur University of Science and Technology Mirpur Pakistan Department of Engineering Manchester Metropolitan University Manchester United Kingdom

In the last couple of decades, autonomous human assistance robots have been enormously attracting the industrial sector. For this purpose, numerous researchers have contributed towards designing efficient and robust human assistance mechanisms. However, their proposed approaches do not provide a cost-effective solution due to the deployment of exorbitant sensors and sophisticated infrastructure. Besides, it was quite challenging for existing human-following robots to track their assigned human companion in different illusional states and luminous conditions while detecting obstacles and taking respective maneuvers (i.e., abrupt turns, etc.). Moreover, self-driving solutions need to take fast and real-time actions to avoid collisions in the designated environment. For this purpose, literature has shown the efficiency of YOLOv3 with respect to providing real-time results in latency-sensitive applications. Hence, to overcome this dilemma, we propose to develop an economically efficient deep learning- based smart labor assistance trolley that uses the YOLO v3 as a core detective deep learning technique with advance and efficient perception and motion planning modules. The perception module succours the autonomous trolley to precisely detect and classify the objects. While the motion planning module uses the specific intended target detection technique to follow the targeted person in crowded environment. These techniques make the autonomous trolley able to take expeditious, meticulous, and conspicuous action in real-time. The labor assistant robot detects and tracks the respective person using YOLOv3. To validate the efficiency of the proposed solution, we have performed a series of experiments considering different test cases. Our proposed work achieved a mean average precision of 0.81%.

关键词： Automation Target tracking Service robots Transportation Object detection Robot sensing systems Real-time systems

来源：评论

学校读者我要写书评

暂无评论

A Novel Approximation for the Spring Loaded Inverted Pendulum Model of Locomotion

A Novel Approximation for the Spring Loaded Inverted Pendulu...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Ali U. Kilic David J. Braun Department of Mechanical Engineering Advanced Robotics and Control Laboratory Center for Rehabilitation Engineering and Assistive Technology Vanderbilt University Nashville Tennessee USA

The Spring-Loaded Inverted Pendulum (SLIP) is one of the simplest models of robot locomotion. SLIP is commonly used to predict the center of mass motion and derive simple control laws for stable locomotion. However, the SLIP model is not integrable, which means that no closed-form relation can be derived to understand how the design and control parameters of the SLIP model affect stable locomotion. There exist a number of different analytical approximations to the SLIP model when considering small step lengths and symmetric steps. In this paper, we present a novel approximation to the SLIP model without relying on the small step length and the symmetric step assumption. The model was found to accurately predict the stability of the SLIP model for large and asymmetric steps and was used to design a controller to stabilize the SLIP model in a couple of steps.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Bearing Fault Diagnosis Based on Multiple Feature Transfer Learning Network 11

Bearing Fault Diagnosis Based on Multiple Feature Transfer L...

引用

11th IEEE Data Driven control and Learning Systems Conference, DDCLS 2022

作者： Qiao, Huanzhang Liu, Lida Tian, Xincheng Liu, Yan Center for Robotics School of Control Science and Engineering Shandong University Jinan250061 China Engineering Research Center of Intelligent Unmanned System Ministry of Education Jinan250061 China Shandong Raone Technology Co. LTD Jinan250012 China

ISBN: (数字)9781665496759

ISBN: (纸本)9781665496759

The intelligent fault diagnosis methods have gained great success, they usually depend on large amounts of labeled data. Unfortunately, it is a realistic problem to collected massive labeled data in actual industrial scene. In addition, even labeled data is easy to collect, the methods trained by these labeled data may not be able to classify unlabeled data obtained under different working states. Inspired by transfer learning, in the paper we propose a deep learning-based fault diagnosis method named multiple feature transfer learning network (MFTLN) to resolve the data distribution difference due to changed working conditions. MFTLN employee multiple extractors to get domain-invariant features and align source domain and target domain by maximum mean discrepancy (MMD) metric. We performed six bearing fault diagnosis transfer learning experiments to verify the superiority of the proposed method, the experimental results indicated that the recognition accuracy of the proposed method is obviously higher than those of the other 5 popular domain adaptive fault diagnosis methods when the target domain data is unlabeled. © 2022 IEEE.

关键词： Failure analysis

来源：评论

学校读者我要写书评

暂无评论

AIRFRAME - Fast prototyping framework for UAVs definition

AIRFRAME - Fast prototyping framework for UAVs definition

引用

International Conference on Unmanned Aircraft Systems (ICUAS)

作者： A. Berra P. J. Sanchez-Cuevas M.A. Trujillo G. Heredia A. Viguria CATEC (Advanced Center for Aerospace Technologies) Sevilla Spain Robotics Vision and Control Group (GRVC) Universidad de Sevilla Seville Spain

Developing a new UAV platform is a long and iterative process that requires a lot of time and effort to be successful. The difficulty of performing a realistic evaluation of system performance during the development process represents a major drawback. As a matter of fact, in most contexts, the first proof of UAVs’ capabilities arrives only during the first flights of the real platform. This may lead, in case of possible issues detected in the platform, to a revaluation of the design, which is not optimal at the very last stage of platform development. To overcome this issue, we propose AIRFRAME, a framework for fast-developing UAV prototypes in simulation to allow for systematic evaluation and analysis of UAV performance during the development process. The developed prototype integrates software and hardware for a better evaluation of the system’s capability and performance at an early stage. The implementation of the framework has succeeded with Gazebo-ROS-Matlab in Docker Environment. It allows high integrability and fast evaluation of multiple UAV designs.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Deep Reinforcement Learning Based on Local GNN for Goal-Conditioned Deformable Object Rearranging

arXiv

引用

arXiv 2023年

Object rearranging is one of the most common deformable manipulation tasks, where the robot needs to rearrange a deformable object into a goal configuration. Previous studies focus on designing an expert system for each specific task by model-based or data-driven approaches and the application scenarios are therefore limited. Some research has been attempting to design a general framework to obtain more advanced manipulation capabilities for deformable rearranging tasks, with lots of progress achieved in simulation. However, transferring from simulation to reality is difficult due to the limitation of the end-to-end CNN architecture. To address these challenges, we design a local GNN (Graph Neural Network) based learning method, which utilizes two representation graphs to encode keypoints detected from images. Self-attention is applied for graph updating and cross-attention is applied for generating manipulation actions. Extensive experiments have been conducted to demonstrate that our framework is effective in multiple 1-D (rope, rope ring) and 2-D (cloth) rearranging tasks in simulation and can be easily transferred to a real robot by fine-tuning a keypoint detector. Copyright © 2023, The Authors. All rights reserved.

关键词： Expert systems

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

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：