检索结果-内蒙古大学图书馆

AHLNet:Adaptive Multihead Structure and Lightweight Feature Pyramid Network for Detection of Live Working in Substations

引用

Machine Intelligence Research 2024年第5期21卷 983-992页

作者： Mengle Peng Xiaoyong Jiang Langyue Huang Zhongyi Li Haiteng Wu Xiaotang Geng Zhejiang University of Science and Technology Hangzhou310023China Zhejiang Key Laboratory of Intelligent Operation and Maintenance Robot Hangzhou311100China

With the increasing demand for power in society,there is much live equipment in substations,and the safety and standardization of live working of workers are facing *** at these problems of scene complexity and object diversity in the real-time detection of the live working safety of substation workers,an adaptive multihead structure and lightweight feature pyramid-based network(AHLNet)is proposed in this study,which is based on ***,we take AH-Darknet53 as the backbone network of YOLOV3,which can introduce an adaptive multihead(AMH)structure,reduce the number of network parameters,and improve the feature extraction ability of the backbone ***,to reduce the number of convolution layers of the deeper feature map,a lightweight feature pyramid network(LFPN)is proposed,which can perform feature fusion in advance to alleviate the problem of feature imbalance and gradient ***,the proposed AHLNet is evaluated on the datasets of 16 categories of substation safety operation scenarios,and the average prediction accuracy MAP_(50)reaches 82.10%.Compared with YOLOV3,MAP_(50)is increased by 2.43%,and the number of parameters is 90 M,which is only 38%of the number of parameters of *** addition,the detection speed is basically the same as that of YOLOV3,which can meet the real-time and accurate detection requirements for the safe operation of substation staff.

关键词： Adaptive multihead structure lightweight feature pyramid substation feature imbalance multiobject detection

来源：评论

学校读者我要写书评

暂无评论

GCN Rewarded Reinforcement Learning for Action Segmentation 24

GCN Rewarded Reinforcement Learning for Action Segmentation

引用

4th International Joint Conference on robotics and Artificial Intelligence, JCRAI 2024

作者： Wang, Liangliang Liu, Jiayao Hu, Xujun Huo, Guanglei Wuhan University of Technology Wuhan China Beijing Jiaotong University Beijing China Institute of Advanced Manufacturing Technology Fujian Key Laboratory of Intelligent Operation and Maintenance Robot Technology Quanzhou China

ISBN: (纸本)9798400710100

Action recognition is a very basic topic for human-robot visual interaction. To enable action recognition for real-world applications, at the core is the problem of action segmentation, which aims to segment action slices from long videos. Existing action segmentation approaches rely heavily on hand-crafted techniques, which are limited on accuracy due to the spatio-temporal complexity of actions across videos. To this end, we formulate action segmentation as a Markov decision process rewarded by observing the action recognition results, and build a reinforcement learning framework to adjust the segmenting procedure iteratively according to the rewards. In particular, to perceive the segmented actions for reinforcement learning, we propose a temporal graph convolutional network (GCN) by representing the context relationships of neighboring action slices, whose states are initialized by considering the consistency of actions, and updated by computing their temporal similarity fusing the tIoU and center distance. Convolutional neural network (CNN) backbone features are used as the nodes of GCN, and the edges of GCN are calculated utilizing the context relationships for spatially and temporally representation. By designing an observation and objective module within the reinforcement learning framework based on GCN representations, we finish the network training for. Finally, we conduct extensive experiments on two public datasets, and by comparisons with some state-of-the-art results, our method is demonstrated. © 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

GaitFormer: Two-Stream Transformer Gait Recognition Using Wearable IMU Sensors in the Context of Industry 5.0

引用

IEEE Sensors Journal 2025年第11期25卷 19947-19956页

作者： Lu, Zhangli Zhou, Huiying Wang, Longqiang Kong, Depeng Lyu, Honghao Wu, Haiteng Chen, Bing Chen, Fuguo Dong, Na Yang, Geng Zhejiang University State Key Laboratory of Fluid Power and Mechatronic Systems School of Mechanical Engineering Hangzhou310058 China Innovation Institute Co. Ltd China Hangzhou Shenhao Technology Key Laboratory of Intelligent Robot for Operation and Maintenance of Zhejiang province Hangzhou China

The rising attention to human-centricity, a core value of Industry 5.0, makes researchers propose various methods to place humans in the center of production, such as the Human Digital Twin (HDT) and gait identity recognition. However, previous HDT applications in the Industry 5.0 era have less consideration of identity recognition, the vital way to link physical signals to personal archives in cyber systems. Herein GaitFormer, a Two-Stream Transformer gait recognition method is proposed for application with the wearable inertial measurement unit (IMU) motion capture system. GaitFormer consists of Temporal and Channel Modules enclosing Gaussian Range Encoding, multi-head Auto-Correlation mechanism, and multi-scale as convolutional neural network (CNN) layers. It outperforms state-of-the-art architectures with extensive experiments over publicly available IMU-based datasets respectively: whuGait and OU-ISIR. Then the study intends to combine the motion capture system and gait identity recognition algorithm to propose an application of personal monitoring in the Human-Cyber-Physical-System (HCPS), which uses GaitFormer to recognize the identity of the user with motion capture system and show their corresponding customized HDT. This work may contribute to making a connection between gait identity recognition and HDT, constructing personal working profiles in the environment of Industry 5.0. © 2001-2012 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Revitalizing Human-robot Interaction: Phygital Twin Driven robot Avatar for China-Sweden Teleoperation

引用

Chinese Journal of Mechanical Engineering 2023年第5期36卷 1-6页

作者： Huiying Zhou Honghao Lv Ruohan Wang Haiteng Wu Geng Yang State Key Laboratory of Fluid Power and Mechatronic Systems School of Mechanical EngineeringZhejiang UniversityHangzhou 310027China Hangzhou Shenhao Technology Co. Ltd.Hangzhou 310000China Zhejiang Key Laboratory of Intelligent Operation and Maintenance Robot Hangzhou 310000China

1 Introduction Digital Twin(DT)is built and maintained in the digital rather physical realm[1,2].Dynamic DT mirrors the exact state of the physical system by means of multidimensional models describing the behavior of physical entity,and sensors providing the real-time coupling to models[3].Te outbreak of COVID-19 pandemic results in speeding up the virtualization and digitization of the physical entity with the support from ***,quarantine and division led by COVID-19 pandemic make human beings’eyes for the physical interaction become more *** physical world and the digital world are becoming interwoven,which indicates a new opportunity towards phygital *** is so often the question:how can I get closer to my distant companion?How can I extend my service or emotional concern to facilitate interaction in spite of space obstruction?Tese questions deserve thinking as a lesson learned from outbreak of COVID-19 pandemic.

关键词： Twin robot closer

来源：评论

学校读者我要写书评

暂无评论

An efficient and scalable semi-supervised framework for semantic segmentation

引用

Neural Computing and Applications 2025年第7期37卷 5481-5497页

作者： Hao, Huazheng Xiao, Hui Xiong, Junjie Dong, Li Yan, Diqun Liang, Dongtai Zhuang, Jiayan Peng, Chengbin Ningbo University Ningbo China Zhejiang Rural Commercial Digital Technology Co. Ltd Hangzhou China Zhejiang Key Laboratory of Intelligent Operation and Maintenance Robot Hangzhou Shenhao Technology Co. Ltd Hangzhou China Ningbo Institute of Materials Technology and Engineering Chinese Academy of Sciences Ningbo China

Semantic segmentation has succeeded remarkably in various applications, such as autonomous vehicles and robotic systems. However, the training process for such techniques necessitates a significant amount of labeled data. Although semi-supervised frameworks can alleviate this issue, advanced approaches typically require multiple baseline models to form a dual model, which is costly in space computation complexity. To relieve the undesired computational cost for systems with precious computation and memory resources, we propose an efficient and scalable semi-supervised learning framework to significantly improve the performance with a few additional parameters concerning the baseline models. This framework includes a pseudo-dual module and a self-rectification module. The overall structure comprises three parts: an encoder, a shallow decoder, and a deep decoder. The deep decoder is connected to a deep layer of the encoder, and the shallow decoder is connected to a shallow layer. As knowledge distillation transfers knowledge from one model to another, the pseudo-dual module can distill knowledge from the ensemble of two decoders to improve the encoder, which can implicitly form a pseudo-dual model. The self-rectification module calculates class-wise likelihoods according to the similarity between features and class prototypes learned from different decoders and rectifies low-confidence pseudo-labels. The effectiveness of such rectification is justified theoretically and numerically. In our experiments with DeepLabV2, our methods outperform others in mIoU by over 1.21% with 1/8 labeled data using the Cityscapes dataset and by 0.38% with 1/8 labeled data using PASCAL VOC 2012 datasets. In most cases, our approach can also save more than 30% of memory costs during training. Nevertheless, the effectiveness of the proposed approach also depends on the quality of pseudo-labels generated by backbone models and may encounter challenges when handling data with highly imbalance

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

A Latent Diffusion Based Image Generation Method for Anomaly Detection 8th

A Latent Diffusion Based Image Generation Method for Anomaly...

引用

8th International Conference on Computing, Control and Industrial Engineering, CCIE 2024

作者： Wang, Liangliang Song, Lian Yang, Zhuo Li, Bingyao Liang, Peidong School of Computer Science and Artificial Intelligence Wuhan University of Technology Wuhan China Wuhan University of Technology Chongqing Research Institute Chongqing China Institute of Advanced Manufacturing Technology Quanzhou China Fujian Key Laboratory of Intelligent Operation and Maintenance Robot Technology Quanzhou China

ISBN: (纸本)9789819769360

The primary issue in visual anomaly detection is lack of sufficient samples, which makes it difficult for real-world applications unless an effective anomalous image generation algorithm is proposed. To this end, this paper formulates a latent diffusion model to generate a variety of anomalous images by adapting the initial latent tensors to the anomaly. Therefore, the diffusion process is speeded up via transferring the image synthesis in a low-dimensional space, with anomalous labels provided. The pseudo-anomaly images and pseudo-labels are then used for data augmentation for anomaly detection based on YOLO-v8, and quantitative and qualitative experiments on NEU-DET demonstrate the applicability and superiority of our approach on precision and diversity. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Anomaly detection Deep learning Diffusion model Image generation

来源：评论

学校读者我要写书评

暂无评论

Enable Intuitive and Immersive Teleoperation: Design, Modeling and Control of a Novel Wearable Exoskeleton 16th

Enable Intuitive and Immersive Teleoperation: Design, Modeli...

引用

16th International Conference on intelligent robotics and Applications, ICIRA 2023

作者： Wang, Ruohan Cui, Xi Lv, Honghao Zhang, Guangyao Wu, Haiteng Yang, Geng State Key Laboratory of Fluid Power and Mechatronic Systems School of Mechanical Engineering Zhejiang University Hangzhou China Key Laboratory of Intelligent Robot for Operation and Maintenance of Zhejiang Province Hangzhou Shenhao Technology Hangzhou China

ISBN: (纸本)9789819964857

Haptic devices with human motion capture function enable natural and flexible teleoperation within unstructured environment. As one of the typical types among haptic devices, wearable exoskeleton allows intuitive and immersive teleoperation. For such a portable device, weight and motion compatibility have a great influence on the potential control performance. This paper presents the development of WIE, the wearable intelligent equipment, a 5-DOF portable exoskeleton that is lightweight and accessible. The mechanism of WIE is designed to suit the majority of human using the adjustable links. Kinematic analysis of it shows a high motion compatibility with human upper limb in the workspace. Furthermore, an incremental motion mapping method is adopted in the joint space to control the slave robot. The pick-and-place experiment is conducted to evaluate the performance of WIE as well as the remote-control strategy. The results indicate that the robot manipulator presents a good following behavior according to the human motions captured by WIE. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.

关键词： Remote control

来源：评论

学校读者我要写书评

暂无评论

Design and Realization of a Multi-DoF robotic Head for Affective Humanoid Facial Expression Imitation 16th

Design and Realization of a Multi-DoF Robotic Head for Affec...

引用

16th International Conference on intelligent robotics and Applications, ICIRA 2023

作者： Li, Jiayan Lv, Honghao Zhang, Nan Wu, Haiteng Yang, Geng State Key Laboratory of Fluid Power and Mechatronic Systems School of Mechanical Engineering Zhejiang University Hangzhou China Key Laboratory of Intelligent Robot for Operation and Maintenance of Zhejiang Province Hangzhou Shenhao Technology Hangzhou China

ISBN: (纸本)9789819964826

As the elderly population continues to expand, there is a corresponding rise in the demand for service robotics in home environments, with a particular focus on humanoid service robots. This paper presents the design of a multi-DoF robotic head based on the Facial Action Coding System (FACS). The head comprises 26 servo motors that drive the independent Action Units on the face, enabling the imitation of facial expressions following the FACS theory. Additionally, a camera is embedded in the eye for visual perception and interaction, while a 6-DOF Stewart platform driven by six servo motors is integrated as the neck part, ensuring the head’s flexibility of movement. Finite element simulation experiments are conducted to validate the robotic facial expression imitation performance of 6 basic expressions including happy, sad, surprise, fear, anger and disgust. The optimized load parameters of the corresponding action units are obtained through simulation, allowing for effective facial expression imitation. The proposed robotic head with facial expression imitation functionality has potential utility for improving human-robot interaction in homecare scenarios and the service quality of the homecare robots. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Eye movements

来源：评论

学校读者我要写书评

暂无评论

Keeping Workers Safe in Electric Working: A robot System for High-Voltage Live operation

Keeping Workers Safe in Electric Working: A Robot System for...

引用

2023 IEEE International Conference on Industrial technology, ICIT 2023

作者： Wang, Longqiang Wang, Ruohan Wu, Haiteng Yang, Geng School of Mechanical Engineering Zhejiang University State Key Laboratory of Fluid Power & Mechanical Systems Hangzhou China Hangzhou Shenhao Technology Co. LTD. Zhejiang Key Laboratory of Intelligent Operation and Maintenance Robot Hangzhou China

ISBN: (纸本)9798350336504

In conventional high-voltage live work, operators are required to work near high-voltage cables, which poses risks such as falling and getting an electric shock. To promise the worker's safety, regional breakout of power is one of the common methods that the power sector would deploy. However, with the development of society, a lot of trouble will be brought about due to the economic loss and social impact caused by regional power outages, which pose higher requirements for the power sector. Hence, it is demanding to carry out a safe and effective method for high-voltage live work without power failure. In such a context, we have developed a dual-arm collaborative robotic system for high-voltage live work. The system includes a human-computer interaction interface and a pair of Omega handles, which allow workers to operate remotely at the control center. In laboratory experiments, the system can accurately complete stripping, trimming, and wiring operations. Also, some actions are pre-programmed inside the control system to realize simple operations automatically. This dual-arm robot collaboration system with a teleoperation function can greatly reduce the labor intensity of workers and ensure their safety. © 2023 IEEE.

关键词： Remote control

来源：评论

学校读者我要写书评

暂无评论

Modeling and Control of Differential-Drive Chassis for a Homecare Assistive robot 32

Modeling and Control of Differential-Drive Chassis for a Hom...

引用

32nd IEEE International Symposium on Industrial Electronics, ISIE 2023

作者： He, Ping Lv, Honghao Wu, Haiteng Yang, Geng Polytechnic Institute State Key Laboratory of Fluid Power and Mechatronic Systems Zhejiang University Hangzhou China Zhejiang University State Key Laboratory of Fluid Power and Mechatronic Systems School of Mechanical Enginerring Hangzhou China Hangzhou Shenhao Technology Co. LTD. Zhejiang Key Laboratory of Intelligent Operation and Maintenance Robot Hangzhou China Zhejiang University Zhejiang Key Laboratory of Intelligent Operation and Maintenance Robot State Key Laboratory of Fluid Power and Mechatronic Systems School of Mechanical Enginerring Hangzhou China

ISBN: (纸本)9798350399714

With the increasing of the aging population, the demand for homecare services is escalating and the homecare assistive robots are becoming increasingly popular. In this paper, a differential mobile chassis design for a homecare assistive robot is presented. The mobile chassis is made of aluminum sheet and profile, and adopts a two-wheel differential model as the kinematic model. The STM32 single chip microcomputer is used as the cerebellum to control chassis movement, and the NVIDIA Jetson Nano Developer Kit is employed as the brain for perception and decision-making. The software architecture is based on the back-fore ground system and the ROS. A prototype is developed and experiments on simultaneous localization and mapping (SLAM) as well as autonomous navigation experiments are conducted to verify the practicality of the chassis design. © 2023 IEEE.

关键词： Chassis

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：