检索结果-内蒙古大学图书馆

Multi-Seam Tracking with a Portable Robotic Welding System in Unstructured Environments

学校读者我要写书评

暂无评论

Research Square

Research Square 2022年

作者： Yu, Shuangfei Guan, Yisheng Yang, Zhi Liu, Chutian Hu, Jiacheng Hong, Jie Zhu, Haifei Zhang, Tao School of Electro-mechanical Engineering Guangdong University of Technology Guangzhou510006 China Department of Physical Intelligence Max-Planck Institute for Intelligent Systems Stuttgart70569 Germany School of Engineering and Computer Science Australian National University CanberraACT2601 Australia

Most welding manufacturing of the heavy industry, such as shipbuilding and construction, is carried out in an unstructured workspace. The term Unstructured indicates the production environment is irregular, changeable and without model. In this case, the changeable workpiece position, workpiece shape, environmental background, and environmental illumination should be carefully considered. Because of such complicated characteristics, the welding is currently being relied on the manual operation, resulting in high cost, low efficiency and quality. This work proposes a portable robotic welding system and a novel seam tracking method. Compared to existing methods, it can cope with more complex general spatial curve weld. Firstly, the tracking pose of the robot is modeled by a proposed dual-sequence tracking strategy. On this basis, the working parameters can be adjusted to avoid robot-workpiece collision around the workpiece corners during the tracking process. By associating the forward direction of the welding torch with the viewpoint direction of the camera, it solves the problem that the weld feature points are prone to be lost in the tracking process by conventional methods. Point cloud registration is adopted to globally locate the multi-segment welds in the workpiece, since the system deployment location is not fixed. Various experiments on single or multiple welds under different environmental conditions show that even if the robot is deployed in different positions, it can reach the starting point of the weld smoothly and accurately track along the welds. © 2022, CC BY.

关键词： Welds

KDCTime: Knowledge distillation with calibration on inceptiontime for time-series classification

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Gong, Xueyuan Si, Yain-Whar Tian, Yongqi Lin, Cong Zhang, Xinyuan Liu, Xiaoxiang School of Intelligent Systems Science and Engineering Jinan University Zhuhai China Department of Computer and Information Science University of Macau China School of Optoelectronics Beijing Institute of Technology Beijing China

Time-series classification approaches based on deep neural networks are easy to be overfitting on UCR datasets, which is caused by the few-shot problem of those datasets. Therefore, in order to alleviate the overfitting phenomenon for further improving the accuracy, we first propose Label Smoothing for InceptionTime (LSTime), which adopts the information of soft labels compared to just hard labels. Next, instead of manually adjusting soft labels by LSTime, Knowledge Distillation for InceptionTime (KDTime) is proposed in order to automatically generate soft labels by the teacher model. At last, in order to rectify the incorrect predicted soft labels from the teacher model, Knowledge Distillation with Calibration for InceptionTime (KDCTime) is proposed, where it contains two optional calibrating strategies, i.e. KDC by Translating (KDCT) and KDC by Reordering (KDCR). The experimental results show that the accuracy of KDCTime is promising, while its inference time is two orders of magnitude faster than ROCKET with an acceptable training time overhead. © 2021, CC BY.

关键词： Deep neural networks

Driving-Torque Self-Adjusted Triboelectric Nanogenerator for Effective Harvesting of Random Wind Energy

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Wang, Yuqi Li, Xiang Yu, Xin Zhu, Jianyang Shen, Ping Wang, Zhong Lin Cheng, Tinghai Beijing Institute of Nanoenergy and Nanosystems Chinese Academy of Sciences Beijing101400 China Institute of Robotics and Intelligent Systems Wuhan University of Science and Technology Wuhan430081 China State Key Laboratory of Internet of Things for Smart City Department of Civil and Environmental Engineering University of Macau Macao999078 China School of Materials Science and Engineering Georgia Institute of Technology AtlantaGA30332-0245 United States

Triboelectric nanogenerators (TENGs), as a new energy technology for distributed power, are used widely in the field of the natural environment energy harvesting. Because the natural energy is random and unstable, dynamic matching between TENGs driving-torque and natural environment energy is fundamental for improving the applications of TENGs. Therefore, a driving-torque self-adjusted triboelectric nanogenerator (SA-TENG) for effective harvesting of random wind energy is developed in this paper. When the external wind speed is unstable, the SA-TENG automatically self-adjusted its driving-torque to dynamically match the wind speed and obtain higher output power. Experiments showed that the SA-TENG can adjust its driving-torque in accordance to the wind speed ranging in 5.0–13.2 m/s, and that, the output peak power can reach 7.69 mW. Under the same conditions, in comparison with a normal TENG, the power growth rate and the highest energy conversion efficiency of the SA-TENG were boosted by more than 4.3 and 12.2 times, respectively;values that are also 3.2 and 6.5 times higher, respectively, than those of an electromagnetic generator. Additionally, the SA-TENG can supply power to sensors for monitoring environment, proving its feasibility as a distributed energy source. © 2022, The Authors. All rights reserved.

关键词： Nanogenerators

Infrared Invisible Clothing: Hiding from Infrared Detectors at Multiple Angles in Real World

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhu, Xiaopei Hu, Zhanhao Huang, Siyuan Li, Jianmin Hu, Xiaolin School of Integrated Circuits Tsinghua University Beijing China Department of Computer Science and Technology Institute for Artificial Intelligence State Key Laboratory of Intelligent Technology and Systems BNRist Tsinghua University Beijing China IDG/McGovern Institute for Brain Research Tsinghua University Beijing China Beijing China

Thermal infrared imaging is widely used in body temperature measurement, security monitoring, and so on, but its safety research attracted attention only in recent years. We proposed the infrared adversarial clothing, which could fool infrared pedestrian detectors at different angles. We simulated the process from cloth to clothing in the digital world and then designed the adversarial "QR code" pattern. The core of our method is to design a basic pattern that can be expanded periodically, and make the pattern after random cropping and deformation still have an adversarial effect, then we can process the flat cloth with an adversarial pattern into any 3D clothes. The results showed that the optimized "QR code" pattern lowered the Average Precision (AP) of YOLOv3 by 87.7%, while the random "QR code" pattern and blank pattern lowered the AP of YOLOv3 by 57.9% and 30.1%, respectively, in the digital world. We then manufactured an adversarial shirt with a new material: aerogel. Physical-world experiments showed that the adversarial "QR code" pattern clothing lowered the AP of YOLOv3 by 64.6%, while the random "QR code" pattern clothing and fully heat-insulated clothing lowered the AP of YOLOv3 by 28.3% and 22.8%, respectively. We used the model ensemble technique to improve the attack transferability to unseen models. Copyright © 2022, The Authors. All rights reserved.

关键词： Thermography (imaging)

Impedance control for a novel composite modular lower-limb hemiplegic exoskeleton

学校读者我要写书评

暂无评论

Impedance control for a novel composite modular lower-limb h...

2021 IEEE International Conference on Real-Time Computing and robotics, RCAR 2021

作者： Bao, Ruoyu Li, Pengbo He, Bailin Wang, Can Wu, Xinyu Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Guangdong Provincial Key Lab of Robotics and Intelligent System Shenzhen518055 China CAS Key Laboratory of Human-Machine Intelligence-Synergy Systems Shenzhen Institute of Advanced Technology University of Chinese Academy of Sciences Department of Mechanical and Automation Engineering Chinese University of Hong Kong College of Mechanical and Electrical Engineering Shaanxi University of Science and Technology Shaanxi Xi'an710021 China

ISBN: (纸本)9781665436786

This paper proposes a control scheme based on impedance control for the smoothing control of the operating movements of a lower-limb hemiplegic exoskeleton robot on the basis of a new composite modular lower-limb hemiplegic ex-oskeleton(CMLHE) structure. The CMLHE has good structural performance and can be used in the rehabilitation of patients with hemiplegia caused by stroke and other diseases. At the same time, the smoothing control by means of impedance control ensures that patients with lower limb hemiplegia can use the CMLHE to achieve smooth walking movements without additional injuries caused by the CMLHE during the rehabilitation training. This paper analyses the performance of this CMLHE robot based on walking movements and demonstrates through kinematic and kinetic modelling and simulation that the impedance control scheme of this CMLHE robot is feasible and effective and can be more widely used in the rehabilitation training of patients with lower limb hemiplegia. © 2021 IEEE.

关键词： Exoskeleton (robotics)

Energy-optimal Three-dimensional Path-following Control of Autonomous Underwater Vehicles under Ocean Currents

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Yang, Niankai Shen, Chao Johnson-Roberson, Matthew Sun, Jing The Department of Naval Architecture and Marine Engineering University of Michigan Ann ArborMI48109 United States The Department of Systems and Computer Engineering Carleton University OttawaON Canada Department of Naval Architecture and Marine Engineering University of Michigan Ann ArborMI48109 United States The Robotics Institute The School of Computer Science Carnegie Mellon University PittsburghPA15213 United States

This paper presents a three-dimensional (3D) energy-optimal path-following control design for autonomous underwater vehicles subject to ocean currents. The proposed approach has a two-stage control architecture consisting of the setpoint computation and the setpoint tracking. In the first stage, the surge velocity, heave velocity, and pitch angle setpoints are optimized by minimizing the required vehicle propulsion energy under currents, and the line-of-sight (LOS) guidance law is used to generate the yaw angle setpoint that ensures path following. In the second stage, two model predictive controllers are designed to control the vehicle motion in the horizontal and vertical planes by tracking the optimal setpoints. The proposed controller is compared with a conventional LOS-based control that maintains zero heave velocity relative to the current (i.e., relative heave velocity) and derives pitch angle setpoint using LOS guidance to reach the desired depth. Through simulations, we show that the proposed approach can achieve more than 13% energy saving on a lawnmower-type and an inspection mission under different ocean current conditions. The simulation results demonstrate that allowing motions with non-zero relative heave velocity improves energy efficiency in 3D path-following applications. Copyright © 2022, The Authors. All rights reserved.

关键词： Ocean currents

RefineMask: Towards high-quality instance segmentation with fine-grained features

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhang, Gang Lu, Xin Tan, Jingru Li, Jianmin Zhang, Zhaoxiang Li, Quanquan Hu, Xiaolin State Key Laboratory of Intelligent Technology and Systems Institute for AI BNRist Department of Computer Science and Technology Tsinghua University SenseTime Research Tongji University Institute of Automation CAS & UCAS

The two-stage methods for instance segmentation, e.g. Mask R-CNN, have achieved excellent performance recently. However, the segmented masks are still very coarse due to the downsampling operations in both the feature pyramid and the instance-wise pooling process, especially for large objects. In this work, we propose a new method called RefineMask for high-quality instance segmentation of objects and scenes, which incorporates fine-grained features during the instance-wise segmenting process in a multi-stage manner. Through fusing more detailed information stage by stage, RefineMask is able to refine high-quality masks consistently. RefineMask succeeds in segmenting hard cases such as bent parts of objects that are over-smoothed by most previous methods and outputs accurate boundaries. Without bells and whistles, RefineMask yields significant gains of 2.6, 3.4, 3.8 AP over Mask R-CNN on COCO, LVIS, and Cityscapes benchmarks respectively at a small amount of additional computational cost. Furthermore, our single-model result outperforms the winner of the LVIS Challenge 2020 by 1.3 points on the LVIS test-dev set and establishes a new state-of-the-art. Code will be available at https://***/zhanggang001/RefineMask. Copyright © 2021, The Authors. All rights reserved.

关键词：

Evaluation of Robustness Metrics for Defense of Machine Learning systems

学校读者我要写书评

暂无评论

Evaluation of Robustness Metrics for Defense of Machine Lear...

International Conference on Military Communications and Information systems (ICMCIS)

作者： J. DeMarchi R. Rijken J. Melrose B. Madahar G. Fumera F. Roli E. Ledda M. Aktaş F. Kurth P. Baggenstoss B. Pelzer L. Kanestad Collaborative Engineering Systems & Aerospace Systems Information Supremacy Royal Netherlands Aerospace Centre NLR Amsterdam NLD Cyber & Information Systems Division Defence Science and Technology Laboratory Portondown GBR Department of Electrical and Electronic Engineering University of Cagliari Cagliari ITA Department of Informatics Bioengineering Robotics and Systems Engineering University of Genoa Genoa ITA Department of Computer Control and Management Engineering Sapienza University of Rome Rome ITA Defence Systems Technologies Division ASELSAN Ankara TUR Fraunhofer Institute for Communication Information Processing and Ergonomics Bonn DEU Swedish Defence Research Agency Cyber Defence and C2 Technology Division Stockholm SWE

In this paper we explore some of the potential applications of robustness criteria for machine learning (ML) systems by way of tangible “demonstrator” scenarios. In each demonstrator, ML robustness metrics are applied to real-world scenarios with military relevance, indicating how they might be used to help detect and handle possible adversarial attacks on ML systems. We conclude by sketching promising future avenues of research in order to: (1) help establish useful verification methodologies to facilitate ML robustness compliance assessment; (2) support development of ML accountability mechanisms; and (3) reliably detect, repel, and mitigate adversarial attack.

关键词：

Adversarial Texture for Fooling Person Detectors in the Physical World

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Hu, Zhanhao Huang, Siyuan Zhu, Xiaopei Sun, Fuchun Zhang, Bo Hu, Xiaolin Department of Computer Science and Technology Institute for Artificial Intelligence State Key Laboratory of Intelligent Technology and Systems BNRist Tsinghua University Beijing China School of Integrated Circuits Tsinghua University Beijing China IDG/McGovern Institute for Brain Research Tsinghua University Beijing China Beijing China

Nowadays, cameras equipped with AI systems can capture and analyze images to detect people automatically. However, the AI system can make mistakes when receiving deliberately designed patterns in the real world, i.e., physical adversarial examples. Prior works have shown that it is possible to print adversarial patches on clothes to evade DNN-based person detectors. However, these adversarial examples could have catastrophic drops in the attack success rate when the viewing angle (i.e., the camera's angle towards the object) changes. To perform a multi-angle attack, we propose Adversarial Texture (AdvTexture). AdvTexture can cover clothes with arbitrary shapes so that people wearing such clothes can hide from person detectors from different viewing angles. We propose a generative method, named Toroidal-Cropping-based Expandable Generative Attack (TC-EGA), to craft AdvTexture with repetitive structures. We printed several pieces of cloth with AdvTexure and then made T-shirts, skirts, and dresses in the physical world. Experiments showed that these clothes could fool person detectors in the physical world. Copyright © 2022, The Authors. All rights reserved.

关键词： Textures