检索结果-内蒙古大学图书馆

Multi-gait snake robot for inspecting inner wall of a pipeline

Biomimetic Intelligence & robotics 2024年第2期4卷 32-41页

作者： Jingwei Liu Man Li Yahui Wang Da Zhao Rui Deng School of Electrical and Information Engineering Beijing University of Civil Engineering and ArchitectureBeijing 100044China Bejing Key Laboratory of Robot Bionics and Function Research Bejing 100044China

In the field of pipeline inner wall inspection,the snake robot demonstrates significant advantages over other inspection *** a simple traveling wave or meandering motion will suffice for inspecting the inner wall of small-diameter pipes,comprehensively and meticulously inspecting the inner wall of large-diameter pipes requires the snake robot to adopt a helical gait that closely adheres to the inner *** review of existing literature indicates that most research and development on the helical gait of snake robots has focused on the outer surface of cylinders,with very few studies dedicated to developing a helical gait specifically for the inspection of the inner wall of ***,in this study,we propose a helical gait that is suitable for the inner wall of pipes and meets the requirements of gas pipeline *** helical gait is designed using the backbone curve ***,we create a mathematical model for a circular helix curve with constant curvature and torsion,ensuring it is applicable to a snake robot prototype in a laboratory *** calculate the joint angles required for two conical spiral curves with variable curvature and torsion,establish a new model,and define the physical significance of the specific *** ensure the feasibility of the proposed gait,we conduct experiments involving meandering and traveling wave motions to verify the communication and control between the host computer and the snake *** upon this foundation,we further validate the mathematical model of the complex helical motion gait through simulation *** findings provide a theoretical basis for realizing helical movement with a real snake robot.

关键词： Pipeline inner wall detection Snake robot Spiral motion Backbone curve Modeling and simulation

来源：评论

学校读者我要写书评

暂无评论

Visual Realism Assessment for Face-Swap Videos 12th

Visual Realism Assessment for Face-Swap Videos

引用

12th International Conference on Image and Graphics, ICIG 2023

作者： Sun, Xianyun Dong, Beibei Wang, Caiyong Peng, Bo Dong, Jing Beijing Key Laboratory of Robot Bionics and Function Research Beijing University of Civil Engineering and Architecture Beijing China Center for Research on Intelligent Perception and Computing Institute of Automation Chinese Academy of Sciences Beijing China

ISBN: (纸本)9783031463044

Deep-learning-based face-swap videos, also known as deepfakes, are becoming more and more realistic and deceiving. The malicious usage of these face-swap videos has caused wide concerns. The research community has been focusing on the automatic detection of these fake videos, but the assessment of their visual realism, as perceived by human eyes, is still an unexplored dimension. Visual realism assessment, or VRA, is essential for assessing the potential impact that may be brought by a specific face-swap video, and it is also important as a quality assessment metric to compare different face-swap methods. In this paper, we make a small step towards this new VRA direction by building a benchmark for evaluating the effectiveness of different automatic VRA models, which range from using traditional handcrafted features to different kinds of deep-learning features. The evaluations are based on a recent competition dataset named DFGC-2022, which contains 1400 diverse face-swap videos that are annotated with Mean Opinion Scores (MOS) on visual realism. Comprehensive experiment results using 11 models and 3 protocols are shown and discussed. We demonstrate the feasibility of devising effective VRA models for assessing face-swap videos and methods. The particular usefulness of existing deepfake detection features for VRA is also noted. The code can be found at https://***/XianyunSun/***. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Visual Realism Assessment for Face-swap Videos

arXiv

引用

arXiv 2023年

Deep-learning-based face-swap videos, also known as deepfakes, are becoming more and more realistic and deceiving. The malicious usage of these face-swap videos has caused wide concerns. The research community has been focusing on the automatic detection of these fake videos, but the assessment of their visual realism, as perceived by human eyes, is still an unexplored dimension. Visual realism assessment, or VRA, is essential for assessing the potential impact that may be brought by a specific face-swap video, and it is also important as a quality assessment metric to compare different face-swap methods. In this paper, we make a small step towards this new VRA direction by building a benchmark for evaluating the effectiveness of different automatic VRA models, which range from using traditional handcrafted features to different kinds of deep-learning features. The evaluations are based on a recent competition dataset named DFGC-2022, which contains 1400 diverse face-swap videos that are annotated with Mean Opinion Scores (MOS) on visual realism. Comprehensive experiment results using 11 models and 3 protocols are shown and discussed. We demonstrate the feasibility of devising effective VRA models for assessing face-swap videos and methods. The particular usefulness of existing deepfake detection features for VRA is also noted. The code can be found at https://***/XianyunSun/***. Copyright © 2023, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Sclera-TransFuse: Fusing Swin Transformer and CNN for Accurate Sclera Segmentation

Sclera-TransFuse: Fusing Swin Transformer and CNN for Accura...

引用

IEEE International Joint Conference on Biometrics (IJCB)

作者： Haiqing Li Caiyong Wang Guangzhe Zhao Zhaofeng He Yunlong Wang Zhenan Sun Beijing Key Laboratory of Robot Bionics and Function Research Beijing University of Civil Engineering and Architecture Beijing P.R China Beijing University of Posts and Telecommunications Beijing P.R China CRIPAC MAIS CASIA Beijing P.R China

Sclera segmentation is a crucial step in sclera recognition, which has been greatly advanced by Convolutional Neural Networks (CNNs). However, when dealing with non-ideal eye images, many existing CNN-based approaches are still prone to failure. One major reason is that due to the limited range of receptive fields, CNNs are difficult to effectively model global semantic relevance and thus robustly resist noise interference. To solve this problem, this paper proposes a novel two-stream hybrid model, named Sclera-TransFuse, to integrate classical ResNet-34 and recently emerging Swin Transformer encoders. Specially, the self-attentive Swin Transformer has shown a strong ability in capturing long-range spatial dependencies and has a hierarchical structure similar to CNNs. The dual encoders firstly extract coarse- and fine-grained feature representations at hierarchical stages, separately. Then a novel Cross-Domain Fusion (CDF) module based on information interaction and self-attention mechanism is introduced to efficiently fuse the multi-scale features extracted from dual encoders. Finally, the fused features are progressively upsampled and aggregated to predict the sclera masks in the decoder meanwhile deep supervision strategies are employed to learn intermediate feature representations better and faster. Experimental results show that Sclera-TransFuse achieves state-of-the-art performance on various sclera segmentation benchmarks. Additionally, a UBIRIS.v2 subset of 683 eye images with manually labeled sclera masks, and our codes are publicly available to the community through https://***/Ihqqq/Sclera-TransFuse.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Object Mobility classification based Visual SLAM in Dynamic Environments

Object Mobility classification based Visual SLAM in Dynamic ...

引用

International Conference on Ubiquitous robots and Ambient Intelligence (URAI)

作者： Huayan Zhang Tianwei Zhang Yang Li Lei Zhang Wanpeng Wang Beijing Key Laboratory of Robot Bionics and Function Research Beijing University of Civil Engineering and Architecture Beijing China Department of Mechano-Informatics the University of Tokyo Tokyo Japan

ISBN: (数字)9781728157153

ISBN: (纸本)9781728157160

Most of the existed visual odometry methods cannot work in dynamic environments since the dynamic objects lead to wrong uncertain feature associations. In this paper, we involved a learning-based object classification front end to recognize and remove the dynamic object, and thereby ensure our ego-motion estimator's robustness in high dynamic environments. Moreover, we newly classify the environmental objects into static, movable and dynamic three classes. This processing not only enables the ego-motion estimation in the dynamic environment but also leads to clean and complete map-ping results. The experimental results indicate that the proposed method outperformed the other state-of-the-art SLAM solutions in both dynamic and static indoor environments.

关键词： Simultaneous localization and mapping Visualization Feature extraction Task analysis Image segmentation Dynamics

来源：评论

学校读者我要写书评

暂无评论

Nursing robot safety path planning based on improved a star algorithm

引用

Journal of Computers (Taiwan) 2019年第3期30卷 282-289页

作者： Yu, Hongjiang Miao, Xingang Wang, Su Hu, Yawei Beijing Key Laboratory of Robot Bionics and Function Research Beijing University of Civil Engineering and Architecture Beijing100044 China Beijing Electric Power Engineering Company Limited Beijing100070 China

The nursing robot independently developed by our research group is large in size and complex in structure, so it needs suitable path planning algorithm when it works in indoor environment. I If the path planning based on the traditional A* algorithm is adopted, the problem of collision with obstacles may occur. This paper proposes an A* algorithm for virtual obstacle points combined with segmented navigation. Improve the cost estimation function and segment the planning and increase the virtual obstacle points, so that the overall path is centered away from obstacles, avoiding the head and foot collision obstacles. Experimental results show that the proposed algorithm is compared to the traditional A* algorithm: the navigation path is centered, the operation time is increased by 35%-47%, the turning point and the cumulative turning angle are reduced by 50%-80%, and the path length is basically unchanged. The proposed algorithm can meet the safety requirements of nursing and lifting robots, and at the same time improve the computing efficiency to some extent. © 2019 Computer Society of the Republic of China. All rights reserved.

关键词： Nursing

来源：评论

学校读者我要写书评

暂无评论

Distributed Teleoperation System for Controlling Heterogeneous robots Based on ROS

Distributed Teleoperation System for Controlling Heterogeneo...

引用

IEEE Workshop on Advanced robotics and its Social Impacts, ARSO

作者： Fei Yuan Lei Zhang Huayan Zhang Dezhong Li Tianwei Zhang Beijing Key Laboratory of Robot Bionics and Function Research Beijing University of Civil Engineering and Architecture Beijing China School of Information Science and Technology University of Tokyo 7-3-1 Hongo Bunkyo-ku Tokyo Japan

In this paper, we propose a distributed teleoperation system for controlling heterogeneous robots, which improve the problem of the communication of different kinds of robots based on the robot operating system (ROS). The system includes modules for massive data management and multi-level communication, which improves the reuse rate of software code and the flexibility of the robot hardware system by adjusting the overall architecture of the system and decoupling of software and hardware. In our experiments, the distributed control system of the heterogeneous robot was used on the distributed communication and control of the wheeled mobile robot and the humanoid robot, and the remote operation of the heterogeneous robot in the wireless network was successfully realized through the keyboard.

关键词：

来源：评论

学校读者我要写书评

暂无评论

DC-link stability analysis for AC drive system with common DC-bus

引用

Journal of Computers (Taiwan) 2017年第4期28卷 148-156页

作者： Wen, Xiaoyan Beijing Key Laboratory of Robot Bionics and Function Research Beijing University of Civil Engineering and Architecture Beijing100044 China

AC drive system with common DC-bus is widely used in industry process, however, because of inductance in rectifier and negative resistance characteristic of inverter-motor system, DC-bus fluctuation often appears. With small signal method and power-fed AC drive test-bed, this paper studies stability of AC drive system with common DC-bus. First, structure of AC drive system with common DC-bus and power-fed test-bed are introduced, and its small signal mathematic model is derived. Then, stability of test-bed is studied from the point of DC-link stability, and system parameter and operation point effect on stability is also discussed. Beyond that, system stability area is compared with ordinary AC drive system. The conclusion shows that common DC-bus system has a priority than ordinary AC drive system without power-fed structure in the view of DC-bus stability. Experiment results validate above analysis conclusions. What's more, the conclusion provides an advice for drive system parameter choice.

关键词： AC motors

来源：评论

学校读者我要写书评

暂无评论

Designation of Laser Rapid Prototyping Machine Based on ACS Motion Controller

Designation of Laser Rapid Prototyping Machine Based on ACS ...

引用

第28届中国控制与决策会议

作者： Eryu Zhang Xingang Miao Beijing University of Civil Engineering and Architecture Beijing Key Laboratory of Robot Bionics and Function Research

ISBN: (纸本)9781467397155

Laser rapid prototyping(LPR) technology has developing fast in recent *** machine's motion program is developed using library function provided by the developer of the motion controller and ran in computer,which increase communication burden of the *** machine designed in this paper select the products of ACS Motion Control of Israel as the motion controller,motor driver and I/O module and they can communicate through Ether CAT *** main motion program developed in ACSPL+language is stored and ran in the controller,which reduce the interaction times between the computer and motion control card,thus improving the precision,reliability and respond speed of the system.

关键词： Laser Rapid Prototyping ACS motion controller ACSPL+ Ether CAT

来源：评论

学校读者我要写书评

暂无评论

A Control System of Human-Computer Interaction Based on Kinect Somatosensory Equipment

A Control System of Human-Computer Interaction Based on Kine...

引用

第28届中国控制与决策会议

作者： Liang Guo Chenxi Liu Xiaoyan Wen Haohua Chen Jianghui Zhang Beijing University of Civil Engineering and Architecture Beijing Key Laboratory of Robot Bionics and Function Research

ISBN: (纸本)9781467397155

The requirement of human-robot interaction has become increasingly high with the improvement of the robot *** paper presents a robot control system which is based on the *** system through the body movements and gestures to control the *** system is divided into two parts:information collection system and action execution *** human action information is collected by Kinect somatosensory equipment,after the PC analysis parse out the appropriate action,the information is transmitted to the actions performed robot through the bluetooth wireless *** will be received by the robot and the robot controller will control the robot to make the corresponding action,this enables it to achieve real-time human imitation.

关键词： Kinect Human-Computer Interaction robot Control System

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：