检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Benz, Thomas Ottaviano, Alessandro Balas, Robert Garofalo, Angelo Restuccia, Francesco Biondi, Alessandro Benini, Luca Integrated Systems Laboratory ETH Zurich Switzerland Department of Electrical Electronic and Information Engineering University of Bologna Italy Department of Computer Science and Engineering UC San Diego San DiegoCA United States Department of Excellence in Robotics & AI Scuola Superiore Sant’Anna Pisa Italy

The increasing demand for heterogeneous functionality in the automotive industry and the evolution of chip manufacturing processes have led to the transition from federated to integrated critical real-time embedded systems (CRTESs). This leads to higher integration challenges of conventional timing predictability techniques due to access contention on shared resources, which can be resolved by providing system-level observability and controllability in hardware. We focus on the interconnect as a shared resource and propose AXI-REALM, a lightweight, modular, and technology-independent real-time extension to industry-standard AXI4 interconnects, available open-source. AXI-REALM uses a credit-based mechanism to distribute and control the bandwidth in a multi-subordinate system on periodic time windows, proactively prevents denial of service from malicious actors in the system, and tracks each manager’s access and interference statistics for optimal budget and period selection. We provide detailed performance and implementation cost assessment in a 12nm node and an end-to-end functional case study implementing AXI-REALM into an open-source Linux-capable RISC-V SoC. In a system with a general-purpose core and a hardware accelerator’s DMA engine causing interference on the interconnect, AXI-REALM achieves fair bandwidth distribution among managers, allowing the core to recover 68.2 % of its performance compared to the case without contention. Moreover, near-ideal performance (above 95 %) can be achieved by distributing the available bandwidth in favor of the core, improving the worst-case memory access latency from 264 to below eight cycles. Our approach minimizes buffering compared to other solutions and introduces only 2.45 % area overhead compared to the original SoC. Copyright © 2023, The Authors. All rights reserved.

关键词： Embedded systems

来源：评论

学校读者我要写书评

暂无评论

Classification-Detection of Medical Images by Visualizing Hidden-Layer Features of a Deep Learning Approach 12th

Classification-Detection of Medical Images by Visualizing Hi...

引用

12th International Conference on computer engineering and Networks, CENet 2022

作者： Tong, Ziyu Deng, Xing Shao, Haijian Wang, Xia School of Computer Jiangsu University of Science and Technology Zhenjiang212003 China School of Automation Key Laboratory of Measurement and Control for CSE Ministry of Education Southeast University Nanjing Jiangsu210096 China School of Information Science and Technology Nantong University Nantong Jiangsu226019 China Department of Electrical and Computer Engineering National University of Singapore Singapore117583 Singapore

ISBN: (纸本)9789811969003

Deep learning algorithms, such as CNN, have been widely used on a variety of computer vision problems due to their powerful capacity to extract nonlinear features for medical picture analysis. However, due to the decreased identification rate for medical images, CNNs continue to fail to offer an interpretable depiction of the hidden-layers’ features. Furthermore, the enormous amount of redundant parameters from the convolutional to fully connected layers will surely add to the computational complexity. This work provides a set of fast deep learning architectures for visualizing and analyzing hidden-layer properties in medical images. The clustered feature map is used to simplify the model parameters and minimize redundancy parameters in order to meet the goal of a streamlined trade-off between model performance and architecture. The testing results on the medial images reveal that 41% of redundant channels may be removed while 100% of performance can be preserved. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Human-in-the-loop Optimization for Adaptive Assist-as-Needed Rehabilitation 12

Human-in-the-loop Optimization for Adaptive Assist-as-Needed...

引用

12th International Conference on CYBER Technology in automation, Control, and Intelligent Systems, CYBER 2022

作者： Li, Ning Yang, Yang Yang, Tie Chen, Wenyuan Wang, Yihan Yu, Peng Wang, Wenxue Xi, Ning Liu, Lianqing Shenyang110016 China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang110016 China University of the Chinese Academy of Sciences Beijing100049 China State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Shenyang10016 China School of Electrical and Electronics Engineering Nanyang Technological University 50 Nanyang Avenue 639798 Singapore Emerging Technologies Institute Department of Industrial and Manufacturing Systems Engineering University of Hong Kong Pokfulam999077 Hong Kong

ISBN: (数字)9781665472678

ISBN: (纸本)9781665472678

Active voluntary participation in robot-assisted rehabilitation promote the recovery of stroke hemiplegia. However, different patients have personalized recovery states, which needs the adaptive control strategy to improve the patient's participation. Therefore, in this study, a Mirror Adaptive Assist-As-Needed (MAAN) scheme is proposed to encourage subjects to participate actively and avoid the slack of the subjects including two modules: Functional Capability Estimation (FCE) and Mirror Adaptation (MAD). Specifically, the first MSFE module can obtain joint functional capability based on fused biological and motional signals. The second MAD module acquires the needed assistance via the estimated torques of FCE based on healthy limb. Through the iteration and collaboration of these two modules, a human-in-the-loop adaptive training is achieved to promote the patient participation. The results demonstrate that the MAAN strategy can generate suitable assistance for the patients to correct abnormal muscle activation patterns and promote active voluntary participation. Compared with other methods, the proposed strategy improves motion stability and consistency with the natural physiological state of healthy limbs. This research can be expected to greatly enhance the rehabilitation and power assistance outcome of exoskeleton robot in more practical applications. © 2022 IEEE.

关键词： Mirrors

来源：评论

学校读者我要写书评

暂无评论

Fault Diagnostic Opportunities for Electromagnetic Coils of Active Magnetic Bearings using Physics-of-Failure Analysis

Fault Diagnostic Opportunities for Electromagnetic Coils of ...

引用

System Reliability and Safety engineering (SRSE), International Conference on

作者： Ruiqi Li Kai Wang Peng Zeng Jiwei Cao State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Shenyang China Key Laboratory of Networked Control Systems Chinese Academy of Sciences Shenyang China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang China University of Chinese Academy of Sciences Beijing China Shenyang Institute of Automation Chinese Academy of Sciences Shenyang China Department of Electrical Engineering Harbin Institute of Technology China

Active magnetic bearings (AMBs) have many advantages over traditional oil bearings due to their non-contact characteristics. They are environmentally friendly solution and have been proven to be highly reliable and available. Electromagnetic coils are vital components of AMB and they are failure prone. In this paper, the failure mode of magnetic bearing coil and its corresponding failure mechanism are reviewed by FMMEA method, and the existing coil fault diagnosis methods are introduced. Future directions for coil insulation health monitoring aiming at the application characteristics of AMBs are also discussed.

关键词：

来源：评论

学校读者我要写书评

暂无评论

UAV Forest Fire Detection based on RepVGG-YOLOv5

UAV Forest Fire Detection based on RepVGG-YOLOv5

引用

IEEE International Conference on robotics and Biomimetics

作者： Kang Song Yueyuan Zhang Bo Lu Wenzheng Chi Lining Sun Department of Automation Engineering School of Mechanical and Electrical Engineering Soochow University Suzhou China Ningbo Institute of Materi-als Technology & Engineering Chinese Academy of Sciences Zhejiang Key Laboratory of Robotics and Intelligent Manufacturing Equipment Technology Ningbo China Robotics and Microsystems Center School of Mechanical and Electrical Engineering Soochow University Suzhou China

ISBN: (纸本)9781665481106

The safety of forest resources is of great im-portance to natural and public safety, and the efficient and accurate detection of forest fires is an issue of close concern. Considering the limitations of traditional forest fire detection methods, such as the limited range of view from watchtowers and insufficient resolution of satellite images, a deep learning-based Unmanned Aerial Vehicle (UAV) fire detection system has been proposed. With its high mobility, images of forest fires can be captured by camera-equipped UAV and transmitted to the ground station in real time. The ground station uses a deep learning object detection algorithm to achieve the recognition of fire. In this paper, we choose the state-of-the-art object detection algorithm YOLOv5 for the UAV forest fire detection scenario, and to further improve the detection performance of YOLOv5, we propose RepVGG-YOLOv5 by drawing on the RepVGG network, a training-time multi-branch topology with an inference-time plain architecture. Then we test the models on a self-built dataset, and the experimental results show that the improved RepVGG-YOLOv5 model outperforms the YOLOv5 model in terms of detection performance and detection speed, which proves the feasibility of this paper's model in the UAV forest fire detection systems and the effectiveness of the model improvement. Furthermore, the great potential of this paper's object detection model for real-time applications of accurate forest fire detection has also been demonstrated.

关键词： Satellites Forestry Object detection Autonomous aerial vehicles Real-time systems Inference algorithms Safety

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

IEEE International Conference on robotics and Biomimetics

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

ISBN: (纸本)9781665481106

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.

关键词： Technological innovation Computational modeling Biomimetics Pose estimation Feature extraction Robustness Data mining

来源：评论

学校读者我要写书评

暂无评论

SVM-based Extended Kalman Filter State Estimation of Biped Robot

SVM-based Extended Kalman Filter State Estimation of Biped R...

引用

2022 IEEE International Conference on robotics and Biomimetics, ROBIO 2022

作者： Gao, Chengzhi Xie, Ye Zhu, Shiqiang Huang, Guanyu Kong, Lingyu Xie, Anhuan Gu, Jianjun Zhang, Dan Research Center for Intelligent Robotics Research Institute of Interdisciplinary Innovation Zhejiang Laboratory Hangzhou311100 China Zhejiang Engineering Research Center for Intelligent Robotics Hangzhou311100 China Zhejiang University Zhejiang Hangzhou311100 China York University North YorkM3J 1P3 Canada Department of Electrical and Computer Engineering Dalhousie University HalifaxNSB3H 4R2 Canada

ISBN: (纸本)9781665481090

At present, more and more attention has been paid to the exploration of biped robots. As the precondition of biped stable walking, the state estimation of robot has also received the attention of many experts. Biped robot has special motion characteristics, has a hybrid dynamic system and belongs to a floating base, so it must obtain the contact information of two feet before realizing the motion state estimation of the robot. The contribution of this paper is to introduce soft support vector machine to detect contact phase. This method based on support vector machine (SVM) eliminates the need for force sensors on two feet because they are vulnerable to the impact between the foot and the contact surface. In addition, precise dynamic modeling is not required. Furthermore, this paper uses SVM to obtain the touchdown information, and proposes a state estimation method using extended Kalman filter to combine motion estimation with IMU data. Finally, the touchdown detection algorithm and the state estimation algorithm are verified through experiments, and the accuracy and feasibility of the SVM method are proved through the touchdown detection experiments. Based on this, the full state estimation of the robot is realized through the state estimation experiments. © 2022 IEEE.

关键词： State estimation

来源：评论

学校读者我要写书评

暂无评论

Integrated Sensing and Communications in Clutter Environment

arXiv

引用

arXiv 2023年

作者： Luo, Hongliang Wang, Yucong Luo, Dongqi Zhao, Jianwei Wu, Huihui Ma, Shaodan Gao, Feifei Department of Automation Tsinghua University Beijing100084 China High-Tech Institute of Xi’an Xi’an710025 China State Key Laboratory of Internet of Things for Smart City the Department of Electrical and Computer Engineering University of Macau 999078 China

In this paper, we propose a practical integrated sensing and communications (ISAC) framework to sense dynamic targets from clutter environment while ensuring users communications quality. To implement communications function and sensing function simultaneously, we design multiple communications beams that can communicate with the users as well as one sensing beam that can rotate and scan the entire space. To minimize the interference of sensing beam on existing communications systems, we divide the service area into sensing beam for sensing (S4S) sector and communications beam for sensing (C4S) sector, and provide beamforming design and power allocation optimization strategies for each type sector. Unlike most existing ISAC studies that ignore the interference of static environmental clutter on target sensing, we construct a mixed sensing channel model that includes both static environment and dynamic targets. When base station receives the echo signals, it first filters out the interference from static environmental clutter and extracts the effective dynamic target echoes. Then a complete and practical dynamic target sensing scheme is designed to detect the presence of dynamic targets and to estimate their angles, distances, and velocities. In particular, dynamic target detection and angle estimation are realized through angle-Doppler spectrum estimation (ADSE) and joint detection over multiple subcarriers (MSJD), while distance and velocity estimation are realized through the extended subspace algorithm. Simulation results demonstrate the effectiveness of the proposed scheme and its superiority over the existing methods that ignore environmental clutter. Copyright © 2023, The Authors. All rights reserved.

关键词： Clutter (information theory)

来源：评论

学校读者我要写书评

暂无评论

Long-Term Time Series Anomaly Detection with Transformer and Variational Autoencoder for Water Quality

SSRN

引用

SSRN 2024年

作者： Lin, Yongze Qiao, Junfei Bi, Jing Yuan, Haitao Zhou, MengChu Beijing Laboratory of Smart Environmental Protection Beijing University of Technology Beijing100124 China Faculty of Information Technology Beijing University of Technology Beijing100124 China School of Automation Science and Electrical Engineering Beihang University Beijing100191 China Department of Electrical and Computer Engineering New Jersey Institute of Technology NewarkNJ07102 United States

In many fields, time series anomaly detection is gaining more and more attention, \textit{e.g.}, air pollution, geological hazards, and network traffic detection. Time series anomaly detection in water quality aims to identify anomalous samples. However, there are several challenges in anomaly detection of water quality time series: 1) Difficulty in obtaining local and global temporal dependencies of time series;2) Difficulty in learning a representation map from time series that enables effective discrimination of anomalies;and 3) Complex nonlinear relationship in water quality time series, making it difficult to detect long-term anomalies. To solve these problems, this work proposes a time series anomaly detection model called Transformer with Variational Autoencoder (TransVAE). First, this work integrates local and global temporal position embedding to improve temporally dependent representation. Second, It adopts a Variational Autoencoder (VAE) to fuse signals and reconstruct their expected distribution. Third, we use a Transformer to process nonlinear relationships in the time series and use residual mechanisms to improve long-term series anomaly detection accuracy. This work adopts two real-world datasets to evaluate the proposed TransVAE, and experiments demonstrate that TransVAE outperforms several state-of-the-art methods in terms of time series anomaly detection accuracy. © 2024, The Authors. All rights reserved.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

Low-cost Mars Terrain Classification System Based on Coarse-grained Annotation

Low-cost Mars Terrain Classification System Based on Coarse-...

引用

IEEE International Conference on robotics and Biomimetics

作者： Jian Zhang Yeheng Chen Shiqiang Zhu Yuehua Li Tian Xie Jason Gu Zhejiang Laboratory Research Center for Intelligent Robotics Research Institute of Interdisciplinary Innovation Hangzhou China Zhejiang Engineering Research Center for Intelligent Robotics Hangzhou China Department of Electrical and Computer Engineering Dalhousie University Halifax NS Canada

ISBN: (纸本)9781665481106

Due to communication latency with remote ground sites, automatic recognition of Mars terrain is essential for the path-planning of rovers. Currently, most vision-based terrain classification require thousands of fine-grained training samples, while the undefined terrains on Mars are difficult to be classified or fine-grained labeled. Actually, most of the terrain categories can only be coarse-grained labeled due to several limitations, such as overlapped sub-regions, blurred borders, etc. To solve this problem, CACMT (Coarse-grained Annotation-based Classification for Mars Terrain) is proposed to generate the global fine-grained classification map from the local coarse-grained data. Specifically, the complete pipeline is decomposed into (i) annotation rules with unique design, (ii) hierarchical feature fusion network for predicting sub-features of terrain (iii) and a generator for outputting dense terrain categories of Mars. Finally, the results of actual data on Mars demonstrate that the terrain sub-features can be successfully recognized and a dense terrain classification map can be generated applying only coarse-grained labeled images.

关键词： Training Image recognition Annotations Biomimetics Pipelines Generators Robots

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：