检索结果-内蒙古大学图书馆

2021 China Automation Congress, CAC 2021

作者： Zhang, Jin Gan, Yahui Wang, Zhengwei Dai, Xianzhong School of Automation Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing China

ISBN: (纸本)9781665426473

Dual-arm collaborative robots have gradually become a research hotspot in the field of robotics due to their unique collaboration and flexibility, and they have a wide range of application prospects especially in the assembly manufacturing industry. However, the existing dual-arm robot lacks autonomy in assembly operations, especially when faced with a large number of randomly placed assembly parts, how to make the robot have independent assembly task decision-making ability is a challenge. In response to this problem, this paper proposes a new dual-arm robot collaborative assembly task sequence planning algorithm, which fully considers the constraints of the dual-arm robot's mechanical structure, kinematics, characteristics of objects to be assembled, interference between objects, etc.. Firstly, The algorithm uses the Monte Carlo method to analyze the working space of the left and right arms of the dual-arm robot, while taking into account the structural constraints of the robot. Then, according to the characteristics of the object to be assembled, the idea of interference sphere is proposed to analyze the interference between objects in the assembly task. The influence of the operation, and finally combined with the movement cost of the dual-arm assembly, the assembly task sequence is effectively planned. The algorithm was verified by simulation using the ABB YuMi dual-arm collaborative robot, and the experimental results proved the feasibility and effectiveness of the algorithm. © 2021 IEEE

关键词： Monte Carlo methods

来源：评论

学校读者我要写书评

暂无评论

DeepTQ: Predictive TSN Switch Queue Length Based on Deep Learning 1

引用

7th Chinese Conference on Swarm Intelligence and Cooperative control, CCSICC 2023

作者： Jia, Yanxin Xu, Long Xiong, Wei Wang, Xin Shang, Zhijun Yuan, Zijun 3onedata Co. Ltd. Shenzhen China Ministry of Education School of Automation and the Key Laboratory of Measurement and Control of Complex Systems of Engineering Southeast University Nanjing China 3onedata Qitong Co. Ltd. Shanghai China University of Chinese Academy of Sciences Beijing China

ISBN: (数字)9789819733361

ISBN: (纸本)9789819733354

With the rapid development of network technology, TSN (Time-Sensitive Networking) is in the stage of rapid development. It needs to ensure the deterministic transmission of time-sensitive flow, further improve the network throughput, and meet the increasing traffic demand. SDN (Software-Defined Networking) can realize the automatic management of network resources by decoupling the control plane and the data plane. Through global network information, it uses efficient route scheduling to achieve load balancing and improve network throughput. However, the prerequisite for route scheduling is that accurately predict the queue length of TSN switch, so that route scheduling can be carried out according to the predicted results. To address the queue length prediction problem of TSN switch, DeepTQ (Predictive TSN Switch Queue Length Based on Deep Learning) is proposed, which predicts the queue length of TSN switches in the next time slot using deep learning methods. It consists of the feature selection phase and the model prediction phase. In DeepTQ, the feature selection phase uses the Spearman rank correlation coefficient to remove redundant features in the datasets. The model prediction phase uses an improved Transformer Encoder and GRU (Gate Recurrent Unit) to construct the prediction model. Simulation results demonstrate that DeepTQ achieves superior performance compared to existing methods and individual components contribute to its overall effectiveness. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Software defined networking

来源：评论

学校读者我要写书评

暂无评论

Seismocardiographic Signals Detection Based on Circulant Singular Spectrum Analysis through Millimeter Wave Radar 6

Seismocardiographic Signals Detection Based on Circulant Sin...

引用

6th International Conference on Signal and Image Processing, ICSIP 2021

作者： Hu, Haiyang Chen, Yuhao Li, Shihua Yang, Jun Sun, Zhenxing Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education School of Automation Nanjing China

ISBN: (纸本)9780738133737

Aiming at the problem that Seismocardiography (SCG) signal detection cannot be promoted due to the bulkiness of the equipment and the SCG containing 7 fiducial points is difficult to extract, this paper proposes a non-contact SCG signal detection algorithm based on the circulant singular spectrum analysis (CiSSA) that runs on a millimeter-wave radar platform. In order to make the detection equipment portable and achieve non-contact detection, this paper uses a millimeter-wave radar working in the 60-64GHz frequency band to collect the signal returned from the surface of the human chest cavity. Then the signal is processed by static clutter elimination and maximum energy distance gate positioning algorithm to improve the signal-to-noise ratio, especially to reduce the interference problem caused by people's tiny movement. For the key problem that the SCG curve extracted by traditional filtering methods are fuzzy and the fiducial points are difficult to identify, the CiSSA method is used to improve the accuracy and completeness of SCG curve extraction. By CiSSA, the signal is embedded into trajectory matrix, and then decomposed into component matrices containing various frequency characteristics. Next, the component matrices are grouped according to the frequency distribution characteristics of the SCG signal, and finally the result of the grouping is reconstructed into a high-precision SCG signal with clear and distinguishable fiducial points. Experiments show that the proposed SCG detection method can accurately extract the SCG curve, clearly distinguish 7 fiducial points of the SCG, and the success rate of detection can reach 97%. This provides the possibility for the popularization of SCG signal detection. © 2021 IEEE.

关键词： Signal detection

来源：评论

学校读者我要写书评

暂无评论

A camera calibration method based on scene and human semantics 37

A camera calibration method based on scene and human semanti...

引用

37th Youth Academic Annual Conference of Chinese Association of Automation, YAC 2022

作者： Shu, Yuan Wang, Yangang Southeast University Chien-Shiung Wu College Nanjing China Southeast University School of Automation Nanjing China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing China

ISBN: (纸本)9781665465366

Camera calibration is an important preliminary for many applications such as 3D reconstruction, motion capture and human-computer interactions. Although great achievements have been made in the past several years, it still remains two main problems. The one is that conventional calibration methods always depend on specific calibration tools (e.g., checkerboard). The other is that the calibration procedure is complex when these methods are used in multi-view large scenarios. In this paper, we propose a camera calibration method which can be applied to large-scale sports scenarios without a special calibration object. To tackle these challenges, our key-idea is to use scene and human semantics detected from RGB images to calibrate the intrinsic and extrinsic parameters of multi-view cameras. We first detect the scene and human keypoints, then the initial camera parameters are calculated with the static scene keypoints and a predefined sports field model. Finally, we combine the scene and human semantics to refine the camera intrinsic and extrinsic parameters. Experimental results demonstrate that our method successfully obtains accurate camera intrinsic and extrinsic parameters for large-scale sports scenarios. © 2022 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

LIFTEDCL: LIFTING CONTRASTIVE LEARNING FOR HUMAN-CENTRIC PERCEPTION 11

LIFTEDCL: LIFTING CONTRASTIVE LEARNING FOR HUMAN-CENTRIC PER...

引用

11th International Conference on Learning Representations, ICLR 2023

作者： Chen, Ziwei Li, Qiang Wang, Xiaofeng Yang, Wankou School of Automation Southeast University China Key Lab of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University Nanjing China Y-tech Kuaishou Technology China Institute of Automation Chinese Academy of Sciences China

Human-centric perception targets for understanding human body pose, shape and ***-training the model on large-scale datasets and fine-tuning it on specific tasks has become a well-established paradigm in human-centric ***, self-supervised learning methods have re-investigated contrastive learning to achieve superior performance on various downstream *** handling human-centric perception, there still remains untapped potential since 3D human structure information is neglected during the task-agnostic *** this paper, we propose the Lifting Contrastive Learning (LiftedCL) to obtain 3D-aware human-centric representations which absorb 3D human structure *** particular, to induce the learning process, a set of 3D skeletons is randomly sampled by resorting to 3D human kinematic *** this set of generic 3D samples, 3D human structure information can be learned into 3D-aware representations through adversarial *** results demonstrate that LiftedCL outperforms state-of-the-art self-supervised methods on four human-centric downstream tasks, including 2D and 3D human pose estimation (0.4% mAP and 1.8 mm MPJPE improvement on COCO 2D pose estimation and Human3.6M 3D pose estimation), human shape recovery and human parsing. © 2023 11th International Conference on Learning Representations, ICLR 2023. All rights reserved.

关键词： Large datasets

来源：评论

学校读者我要写书评

暂无评论

Multi Scale Attention Network for Crowd Counting 21

Multi Scale Attention Network for Crowd Counting

引用

5th International Conference on Computer Science and Application engineering, CSAE 2021

作者： Yang, Xiangpeng Lu, Xiaobo Key Laboratory of Measurement and Control of Complex Systems of Engineering School of Automation Southeast University Ministry of Education Nanjing China

ISBN: (纸本)9781450389853

Reasonable management and control of extra crowded scenes have become a hot topic in recent years. Counting people from density map generated from the object location annotations is an effective way to analyze crowd information and control crowds in severely congested scenes. In this paper, we propose a novel endto- end crowd counting method called MSANet for crowd counting. MSANet consists of the VGG16 backbone as the fronted part, two branches as the back-end part, including the attention map extractor to predict crowd states (means with people or not), and density map branch to regress the density map. What is more, to obtain high-resolution density map, we combine different scale maps from the front part to the back-end part. On the design of the loss function, to enhance the resolution of the predicted map and its structural similarity to ground truth, we proposed a new loss function for crowd counting. The test result based on the public dataset ShanghaiTech and Subway Crowd Counting Dataset supported by the Nanjing Metro demonstrates the effectiveness of our method.. © 2021 Association for Computing Machinery. All rights reserved.

关键词： Statistical tests

来源：评论

学校读者我要写书评

暂无评论

Research on Drag control Algorithm of Cooperative Robot Based on Torque Compensation

Research on Drag Control Algorithm of Cooperative Robot Base...

引用

2021 China Automation Congress, CAC 2021

作者： Zhu, Dezhi Gan, Yahui Dai, Xianzhong School of Automation Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing China

ISBN: (纸本)9781665426473

At present, the teaching methods of collaborative robot drag based on joint current feedback mainly include integral and partial dynamics model, both of which can make robots dragged. However, there are few researches aimed at the advantages and disadvantages of the drag effect and the specific application requirements of the systematic comparative study. Based on the above problems, two zero-force control methods are used on the same 6 degree-of-freedom(DOF) cooperative robot in this paper. And method of torque compensation is improved by parameter identification. Firstly, both of the two methods need to model and analyse the robotic arm. However, processes of the method based on integral dynamics model are modeling, excitation trajectory, filtering, identification and vertification. The zero-force control based on partial dynamics model mainly focuses on the compensation of gravity torque and static friction torque. The gravity term in the dynamic equation is adopted by modeling. The difference in this part is that the parameters of gravity moment function are identified, that is, the weight moment is calculated under the condition of the mass and center of mass deviation are unknown. Finally, NaiWei robot which is a six-joints cooperative robot in China is taken as the experimental object to test and compare. The experimental results show that method of partial dynamics model needs smaller force than integral dynamics model. And the comparison of drag power can be displayed by sensor. © 2021 IEEE

关键词： Force control

来源：评论

学校读者我要写书评

暂无评论

Long-Term Unit Commitment Considering the Flexibility of Maintenance and Trading Plan 4

Long-Term Unit Commitment Considering the Flexibility of Mai...

引用

4th IEEE International Conference on Renewable Energy and Power engineering, REPE 2021

作者： Li, Yuanjing Jiang, Jingxiao Bao, Yueshuang Liu, Xinyuan Wang, Ying Zhang, Kaifeng Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing China State Grid Shanxi Electric Power Research Institute Taiyuan China Southeast University Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing China

ISBN: (纸本)9781665443715

In recent years, with the rapid development of power market, higher requirements are put forward for the coordination of long-Term unit commitment scheduling and maintenance plan. In a meanwhile, trading planning and maintenance planning are two interrelated and coupled applications, and the joint optimization between them should be considered and studied. In this paper, a long-Term unit commitment model simultaneously adding the constrains of unit maintenance and trading plan is established, with their flexibility taken into consideration. Then, in order to solve the problem of complex and large-scale calculation in long-Term unit commitment scheduling, this paper proposes a peak-valley load based model. After that, based on a 6-bus system, the simulation experiment is carried out, and the effects of maintenance and trading plan on unit commitment scheduling and unit output is analyzed. The simulation results verify the effectiveness of the proposed model. © 2021 IEEE.

关键词： Maintenance

来源：评论

学校读者我要写书评

暂无评论

Video Based Fall Detection Using Human Poses 9th

Video Based Fall Detection Using Human Poses

引用

9th CCF Conference on Big Data, BigData 2021

作者： Chen, Ziwei Wang, Yiye Yang, Wankou School of Automation Southeast University Nanjing210096 China Key Lab of Measurement and Control of Complex Systems of Engineering Ministry of Education Southeast University Nanjing210096 China

ISBN: (纸本)9789811697081

Video based fall detection accuracy has been largely improved due to the recent progress on deep convolutional neural networks. However, there still exist some challenges, such as lighting variation, complex background, which degrade the accuracy and generalization ability of these approaches. Meanwhile, large computation cost limits the application of existing fall detection approaches. To alleviate these problems, a video based fall detection approach using human poses is proposed in this paper. First, a lightweight pose estimator extracts 2D poses from video sequences, and then 2D poses are lifted to 3D poses. Second, we introduce a robust fall detection network to recognize fall events using estimated 3D poses, which increases respective field and maintains low computation cost by dilated convolutions. The experimental results show that the proposed fall detection approach achieves a high accuracy of 99.83% on large benchmark action recognition dataset NTU RGB+D and real-time performance of 18 FPS on a non-GPU platform, 63 FPS on a GPU platform. © 2022, Springer Nature Singapore Pte Ltd.

关键词： Fall detection

来源：评论

学校读者我要写书评

暂无评论

Information Entropy Guided Height-aware Histogram for Quantization-friendly Pillar Feature Encoder

arXiv

引用

arXiv 2024年

作者： Zhou, Sifan Yuan, Zhihang Yang, Dawei Zhao, Ziyu Hu, Xing Shi, Yuguang Lu, Xiaobo Wu, Qiang School of Automation Southeast University the Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education Nanjing210096 China Houmo AI Nanjing210046 China

Real-time and high-performance 3D object detection plays a critical role in autonomous driving and robotics. Recent pillar-based 3D object detectors have gained significant attention due to their compact representation and low computational overhead, making them suitable for onboard deployment and quantization. However, existing pillar-based detectors still suffer from information loss along height dimension and large numerical distribution difference during pillar feature encoding (PFE), which severely limits their performance and quantization potential. To address above issue, we first unveil the importance of different input information during PFE and identify the height dimension as a key factor in enhancing 3D detection performance. Motivated by this observation, we propose a height-aware pillar feature encoder, called PillarHist. Specifically, PillarHist statistics the discrete distribution of points at different heights within one pillar with the information entropy guidance. This simple yet effective design greatly preserves the information along the height dimension while significantly reducing the computation overhead of the PFE. Meanwhile, PillarHist also constrains the arithmetic distribution of PFE input to a stable range, making it quantization-friendly. Notably, PillarHist operates exclusively within the PFE stage to enhance performance, enabling seamless integration into existing pillar-based methods without introducing complex operations. Extensive experiments show the effectiveness of PillarHist in terms of both efficiency and performance. © 2024, CC BY-NC-ND.

关键词： Object detection

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：