检索结果-内蒙古大学图书馆

Chinese control and Decision Conference, CCDC

作者： Xinxiao Liu Chuanjiang Li Department of Control Science and Engineering Harbin Institute of Technology Harbin China

ISBN: (纸本)9781665478977

In this paper, we focus on the problem of surrounding control cooperatively with collision avoidance for uncertain multiple Euler-Lagrange systems with external disturbances. The distributed surrounding control algorithm with collision avoidance is constructed. Firstly, by employing the potential functions to prevent the collision among target and agents, and a robust continuous term with adaptive variation gain is added to decrease the effects of external interference in unknown range. Secondly, in order to handle the nonlinear dynamics, the estimation for uncertain terms is used in the controller design. In addition, the distributed surrounding control algorithm with collision avoidance is proposed such that all the agents can enclose the static target. Finally, an example illustrates the effectiveness of the proposed protocol.

关键词： Satellites Protocols Heuristic algorithms Decentralized control Estimation Interference Nonlinear dynamical systems

来源：评论

学校读者我要写书评

暂无评论

Spatial Transformer Network with Transfer Learning for Small-scale Fine-grained Skeleton-based Tai Chi Action Recognition

arXiv

引用

arXiv 2022年

作者： Yuan, Lin He, Zhen Wang, Qiang Xu, Leiyang Ma, Xiang Department of Control Science and Engineering Harbin Institute of Technology Harbin China

Human action recognition is a quite hugely investigated area where most remarkable action recognition networks usually use large-scale coarse-grained action datasets of daily human actions as inputs to state the superiority of their networks. We intend to recognize our small-scale fine-grained Tai Chi action dataset using neural networks and propose a transfer-learning method using NTU RGB+D dataset to pre-train our network. More specifically, the proposed method first uses a large-scale NTU RGB+D dataset to pre-train the Transformer-based network for action recognition to extract common features among human motion. Then we freeze the network weights except for the fully connected (FC) layer and take our Tai Chi actions as inputs only to train the initialized FC weights. Experimental results show that our general model pipeline can reach a high accuracy of small-scale fine-grained Tai Chi action recognition with even few inputs and demonstrate that our method achieves the state-of-the-art performance compared with previous Tai Chi action recognition methods. Copyright © 2022, The Authors. All rights reserved.

关键词： Large dataset

来源：评论

学校读者我要写书评

暂无评论

Automatic dataset generation for specific object detection

arXiv

引用

arXiv 2022年

作者： Lin, Xiaotian Xu, Leiyang Wang, Qiang Department of Control Science and Engineering Harbin Institute of Technology Harbin China

In the past decade, object detection tasks are defined mostly by large public datasets. However, building object detection datasets is not scalable due to inefficient image collecting and labeling. Furthermore, most labels are still in the form of bounding boxes, which provide much less information than the real human visual system. In this paper, we present a method to synthesize object-in-scene images, which can preserve the objects’ detailed features without bringing irrelevant information. In brief, given a set of images containing a target object, our algorithm first trains a model to find an approximate center of the object as an anchor, then makes an outline regression to estimate its boundary, and finally blends the object into a new scene. Our result shows that in the synthesized image, the boundaries of objects blend very well with the background. Experiments also show that SOTA segmentation models work well with our synthesized data. Copyright © 2022, The Authors. All rights reserved.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

SkeletonCLIP: Recognizing Skeleton-based Human Actions with Text Prompts

SkeletonCLIP: Recognizing Skeleton-based Human Actions with ...

引用

International Conference on Systems and Informatics (ICSAI)

作者： Lin Yuan Zhen He Qiang Wang Leiyang Xu Xiang Ma Department of Control Science and Engineering Harbin Institute of Technology Harbin China

ISBN: (纸本)9781665474498

Human action recognition has been a hot research for decades, and mainstream supervised frameworks include a feature extraction backbone and a softmax classifier to predict daily human actions. When the number of classes applied to the dataset changes, we must retrain the classifier on the well-trained backbone. This pipeline restricts the generalization and transfer ability of the model due to an extra training period. Moreover, replacing action labels with simple number labels discards useful semantic information and can only receive a meaningless classifier at last. In this work, we present a model SkeletonCLIP for skeleton-based human action recognition. We add an alternative text encoder to extract semantic information from labels while keeping the original sequence encoder. We use dot production to measure the similarities of sequence-text pairs in place of traditional classifier head and cross-entropy loss. Experiments from three human action datasets show that our framework can reach a higher recognition accuracy with the help of semantic information when training the network from scratch. The code has been shown at eunseo-v/SkeletonCLIP.

关键词： Training Codes Text recognition Semantics Pipelines Production Feature extraction

来源：评论

学校读者我要写书评

暂无评论

A Study of Electromagnetic Acoustic Emission technology in Non-destructive Testing

A Study of Electromagnetic Acoustic Emission Technology in N...

引用

Sensing, Measurement, Communication and Internet of Things Technologies (SMC-IoT), International Conference on

作者： Yongqi Chang Yi Shen Xin Zhang Shuzhi Song Zhenyu Zhao Huamin Jie Department of Control Science and Engineering Harbin Institute of Technology Harbin China School of Electrical and Electronic Engineering Nanyang Technological University Singapore

ISBN: (数字)9798350373257

ISBN: (纸本)9798350373264

Electromagnetic acoustic emission (EMAE), an emerging and highly promising technology, has gained prominence for its ability to provide profound insights into material integrity evaluation while preserving the specimen integrity. Certain scholars have achieved successful detection of defect in aluminum specimens through the utilization of EMAE technology. However, a comprehensive exploration encompassing the principles, developments, advantages, and limitations pertaining to EMAE remains conspicuously absent within the extant literature. To address aforementioned concern, this paper presents a detailed study on the application of EMAE technology in non-destructive testing (NDT). It categorizes the findings on EMAE based on the distinct electromagnetic interaction mechanisms in ferromagnetic and non-ferromagnetic materials. Additionally, it elucidates the potential of EMAE technology in the detection and identification of rail defects.

关键词： Rails Electric potential Aluminum Acoustic emission Ferromagnetic materials Sensors Internet of Things

来源：评论

学校读者我要写书评

暂无评论

AN EFFICIENT FRAMEWORK FOR FEW-SHOT SKELETON-BASED TEMPORAL ACTION SEGMENTATION

arXiv

引用

arXiv 2022年

作者： Xu, Leiyang Wang, Qiang Lin, Xiaotian Yuan, Lin Department of Control Science and Engineering Harbin Institute of Technology Harbin China

Temporal action segmentation (TAS) aims to classify and locate actions in the long untrimmed action sequence. With the success of deep learning, many deep models for action segmentation have emerged. However, few-shot TAS is still a challenging problem. This study proposes an efficient framework for the few-shot skeleton-based TAS, including a data augmentation method and an improved model. The data augmentation approach based on motion interpolation is presented here to solve the problem of insufficient data, and can increase the number of samples significantly by synthesizing action sequences. Besides, we concatenate a Connectionist Temporal Classification (CTC) layer with a network designed for skeleton-based TAS to obtain an optimized model. Leveraging CTC can enhance the temporal alignment between prediction and ground truth and further improve the segment-wise metrics of segmentation results. Extensive experiments on both public and self-constructed datasets, including two small-scale datasets and one large-scale dataset, show the effectiveness of two proposed methods in improving the performance of the few-shot skeleton-based TAS task. Copyright © 2022, 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

Research on the Flux-weakening Capability of the Reverse-salient Permanent Magnet Synchronous Motor

Research on the Flux-weakening Capability of the Reverse-sal...

引用

International Conference on Electrical Machines and Systems (ICEMS)

作者： Xiaokun Zhao Cheng Zhang Weinan Wang Baoquan Kou Liangkuan Zhu College of Computer and Control Engineering Northeast Forestry University Harbin China Department of Electrical and Automation Harbin Institute of Technology Harbin China

Permanent magnet synchronous motors (PMSMs) offer the benefits of high torque density and a superior power factor. Nevertheless, the challenge lies in the inherent difficulty of adjusting the permanent magnet flux. To address this issue, this paper analyzes the influence of electromagnetic parameters on the constant-power speed range (CPSR). First, the analysis focuses on the influence of d-axis inductance and saliency ratio, aiming to establish a design method for achieving a wide CPSP in PMSMs. Second, based on the electromagnetic parameter requirements, a reverse-salient PMSM is proposed and its working principle is explained. Finally, the electromagnetic characteristics, including torque and the CPSP, are determined through the application of the finite element method (FEM). The obtained results confirm that the reverse-salient PMSM achieves a CPSP of 7:1, thus validating the accuracy of the theoretical analysis.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Fast On board On orbit Servicing Task Planning for GEO Targets

引用

Yuhang Xuebao/Journal of Astronautics 2022年第11期43卷 1454-1465页

作者： Xu, Hang Liang, Weikui Liu, Lujiang Ma, Guangfu Song, Bin Department of Control Science and Engineering Harbin Institute of Technology School of Astronautics Harbin150001 China Aerospace System Engineering Shanghai Shanghai201109 China

Based on the auction algorithm and prior knowledge, a novel improved heuristic auction algorithm (IHAA) is proposed to solve the many-to-many on orbit service mission planning problem in geosynchronous orbit (GEO). Firstly, a representation of the on orbit servicing planning problem based on Lambert transfer is established, and a fast fuel optimal estimation strategy is designed for one to one servicing. Secondly, for the mission planning problem considering J 2 perturbation, a fast solution process is proposed based on auction algorithm. Specifically, to determine the task list of each round in the auction algorithm, a heuristic strategy of target priority ranking based on prior knowledge of Lambert transfer is proposed, which helps determine the order of tasks. In addition, in order to reduce the possibility of allocation failure due to the low servicing capacity of some spacecraft, a re auction mechanism based on the redistribution of failed tasks is also added. Finally, the proposed algorithm is compared with the genetic algorithm and simulated annealing algorithm, the results show that IHAA can quickly give relatively high fitness planning results, and can be used for on board mission planning in the face of emergencies or provide initial value of the optimization algorithm to improve the planning efficiency. © 2022 China Spaceflight Society. All rights reserved.

关键词： Simulated annealing

来源：评论

学校读者我要写书评

暂无评论

Distributed interval estimation for continuous-time linear systems: A two-step method ⋆

引用

IFAC-PapersOnLine 2023年第2期56卷 8500-8505页

作者： Zhenhua Wang Jiahui Zhang Zonglin Liu Weixin Han Mouquan Shen School of Astronautics Harbin Institute of Technology Harbin 150001 China Control and System Theory EECS University of Kassel Kassel Germany Northwestern Polytechnical University Xian P. R. China College of Electrical Engineering and Control Science Nanjing Tech University Nanjing China

This paper proposes a two-step distributed interval estimation method for continuous-time linear time-invariant (LTI) systems subject to unknown but bounded disturbance and measurement noise. First, a network of local observers is designed to obtain point-wise estimation. Second, interval estimation is obtained based on the point-wise estimation and peak-to-peak analysis. The conditions of the observer design and the peak-to-peak analysis are formulated as linear matrix inequalities, which can be solved efficiently. Compared with the distributed interval observers designed using the monotone system theory, the proposed method has less restrictive design conditions and hence broader application scopes. Moreover, the parameter matrices in the proposed method can be optimized to obtain tight interval estimation. A numerical simulation is given to illustrate the performance of the proposed method.

关键词： Continuous time linear system distributed observers interval estimation

来源：评论

学校读者我要写书评

暂无评论

Combined Spacecraft Attitude Tracking control with Prescribed Performance via A Fully Actuated System Approach

Combined Spacecraft Attitude Tracking Control with Prescribe...

引用

Chinese control Conference (CCC)

作者： Guang-Quan Duan Guo-Ping Liu Yan-Ming Fu Center for Control Theory and Guidance Technology Harbin Institute of Technology Harbin P. R. China Department of Electronic Engineering Southern University of Science and Technology Shenzhen P. R. China

In this paper, a fully-actuated system prescribed performance controller is proposed to achieve attitude tracking control of the combined spacecraft. In order to make the combined spacecraft satisfy the required steady-state and transient performance in the attitude tracking control process, a fully actuated system controller with prescribed performance is proposed. The attitude dynamics model of the combined spacecraft is transformed into a tracking error function. Furthermore, the controller of the fully actuated system is designed according to the error function to realize the attitude tracking control of the combined spacecraft. Lyapunov function is introduced to prove the stability of the system. With the presented way of designing fully actuated system prescribed performance controller, the numerical simulation results verify the effectiveness of the proposed method.

关键词：

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：