检索结果-内蒙古大学图书馆

第33届中国控制与决策会议

作者： Siyi Du Fei Wang Guilin Zhou Jiaqi Li Lintao Yang Dongxu Wang Faculty of Robot Science and Engineering Northeastern University

Based on human-computer shared control,this paper introduces a novel robotic manipulation fashion combining computer vision and brain-computer interface(BCI).Designed for disabled groups,the intelligent wheelchair with our proposed method exhibits the precise robotic manipulation ability but also the human decision-making capabilities,which will bring better life quality for the *** overall pipeline includes three parts:asynchronous brain-computer interface based on steady-state visual evoked potential(SSVEP),vision detection with deep network and robotic manipulation of UR5 ***,first,the user receives the periodic visual stimulation with different frequencies and then electroencephalography(EEG) signals of the user are collected by EEG ***,we preprocess the EEG signals and extract feature *** judge the frequency of the stimulus signals received by the user,the canonical correlation analysis(CCA) algorithm is used to fit and compare it with the standard EEG *** our work,the signals with different frequencies corresponds to different types of objects item by ***,we apply the off-theshelf vision detection algorithm,Mask-RCNN,to output the position of the object corresponding to the detected EEG in the image ***5 robot arm plan manipulation path according to the position of objects transferred by robot operating system(ROS).Extensive experiments show that our method can achieves performance with more than 90%accuracy and the user can control the robot arm to grab the expected object accurately through BCI.

关键词： Intelligent wheelchair robotic manipulation BCI SSVEP human-machine shared control

来源：评论

学校读者我要写书评

暂无评论

Cooperative obstacle avoidance method of multiple unmanned underwater vehicles based on improved artificial potential field method

Cooperative obstacle avoidance method of multiple unmanned u...

引用

第35届中国控制与决策会议

作者： HongLi Xu Kuo Luan BenQing Jia HaiTao Gu dept.Faculty of Robot Science and Engineering Northeastern University dept.State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences

Aiming at the obstacle avoidance problem of multiple unmanned underwater vehicles(UUV) formation in different obstacle avoidance scenarios,the collaborative collision avoidance algorithm of multiple UUV formation was *** terms of obstacle avoidance algorithm,an algorithm combining artificial potential field method and consistent formation control algorithm is proposed.A variety of obstacle scenes are set by using the information of forward looking sonar,and different weights are set for the fusion of control instructions of the two *** problems of formation recovery time and formation error caused by formation "grouping" in the course of UUV obstacle avoidance are ***,in the underwater simulation platform,the effectiveness and advantages of the collaborative obstacle avoidance algorithm in multiple scenes are verified by experiments.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An Improved Indoor Navigation Scheme Based on Vision-Language Localization

An Improved Indoor Navigation Scheme Based on Vision-Languag...

引用

第33届中国控制与决策会议

作者： Ziheng Xu Zixi Jia Xuegang Zhou Huan Wen Yanan Li Faculty of Robot Science and Engineering Northeastern University

This paper proposes an improved indoor localization and navigation method based on image matching and text *** with traditional indoor localization methods based on WIFI,Bluetooth or laser,the improved method is mainly realized by image matching in computer vision and text extraction in natural language *** improved bidirectional A*algorithm is adopted to realize the optimal path planning,and the specific location of indoor localization and path planning diagrams are displayed through an *** results show that this method which can realize indoor location determination in real scenes,is convenient to use,and has low *** can overcome the problems of high hardware requirements,high cost,and poor transferability in the traditional indoor navigation field in terms of localization.

关键词： Computer Vision Natural Language Processing Indoor Localization Bidirectional A*Algorithm

来源：评论

学校读者我要写书评

暂无评论

FCOS-Lite:An Efficient Anchor-free Network for Real-time Object Detection

FCOS-Lite:An Efficient Anchor-free Network for Real-time Obj...

引用

第33届中国控制与决策会议

作者： Shuai Liu Jianning Chi Chengdong Wu Faculty of Robot Science and Engineering Northeastern University

Fully Convolutional One-Stage detector(FCOS) is a simple and strong anchor-free ***,the outstanding performance comes at the cost of long inference *** this paper,a more efficient anchor-free detector FCOS-Lite is proposed to speed up the inference time of ***-Lite improves the three components of FCOS,namely the backbone,neck and *** first apply an efficient attention module in the backbone to extract more important semantic ***,we propose an adaptive feature-fusion module in the neck to detect small objects ***,we use some strategies in the head of FCOS-Lite to reduce the *** these methods on the FCOS-Lite,we achieve a speed-accuracy balance on the MS COCO ***-Lite's inference time is almost halved based on the FCOS with a slight accuracy drop.(33.2% mAP at 24 ms for FCOS-Lite compared to 37.4% mAP at 44 ms for FCOS on COCO).FCOS-Lite suits for the real-time object detection,improving both accuracy and efficiency of the real-time detector YOLOv3 and FCOS-mobileNet(31.0% mAP at 29 ms for YOLOv3 and 30.9% mAP at 27 ms for FCOS-mobileNet on COCO).

关键词： Object detection anchor free Efficient FCOS-Lite

来源：评论

学校读者我要写书评

暂无评论

Costmap Construction and Pseudo-Lidar Conversion Method of Mobile robot Based on Monocular Camera

Costmap Construction and Pseudo-Lidar Conversion Method of M...

引用

第33届中国控制与决策会议

作者： Haoran Fu Qibin Chen Zhenguo Chen Shiguang Wen Faculty of Robot Science and Engineering Northeastern University

With the development of robotics and computer vision technology,monocular cameras with low cost advantage are widely used in mobile *** the research of mobile robot obstacle avoidance competition based on ROS and monocular RGB camera,this paper does not directly use monocular camera for navigation and obstacle avoidance,but proposes a novel costmap creation and pseudo-Lidar transformation method based on monocular *** main work of this paper is to process monocular vision images,convert the image coordinate system to map coordinate system through camera calibration,display obstacle information on the costmap,extract obstacle edges and publish them in Lidar data *** paper realizes the conversion of monocular vision to map and Lidar,completes the environment perception and mapping of mobile robots,and makes it possible to use Lidar-based navigation algorithms and toolkits.

关键词： Monocular Vision Costmap Pseudo-Lidar ROS Mobile robot

来源：评论

学校读者我要写书评

暂无评论

A Novel Bottom-Up Semi-Supervised Learning Framework for Salient Object Detection 11

A Novel Bottom-Up Semi-Supervised Learning Framework for Sal...

引用

11th IEEE Annual International Conference on CYBER Technology in Automation, Control, and Intelligent Systems, CYBER 2021

作者： Pang, Yu Wu, Hao Wu, Yunhe Wu, Chengdong Zhou, Wei Yu, Xiaosheng Northeastern University Faculty of Robot Science and Engineering Shenyang110169 China Australian Centre for Field Robotics University of Sydney Sydney Australia College of Information Science and Engineering Northeastern University Shenyang110819 China College of Computer Science Shenyang Aerospace University Shenyang110136 China

ISBN: (纸本)9781665425278

Salient object detection, which aims at locating the most important object in a scene (or image), has been extensively studied in various tasks, such as robot vision. In this paper, we present an effective salient object detection framework based on a novel bottom-up semi-supervised learning algorithm, which obviously outperforms the existing works in complex scenes. Given an input image, it is firstly segmented into a fixed number of non-overlapping image patches as basic units. A novel segmentation-based sampling method is developed to select a subset of all image patches as training samples. Then, all samples are divided into labeled and unlabeled groups based on multiple prior cues. The labels of all the unlabeled data are inferred by a novel label propagation mechanism. As a result, a complete training set can be obtained and used to train a classifier to classify all image patches into salient object and background. In addition, we also use neighbor-constraint smoothness function to further boost the saliency map. We compare the proposed method with the state-of-the-art approaches on two datasets. Experimental results demonstrate the effectiveness and superiority of the proposed method. © 2021 IEEE.

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Motion-to-Matching: A Mixed Paradigm for 3D Single Object Tracking

arXiv

引用

arXiv 2023年

作者： Li, Zhiheng Lin, Yu Cui, Yubo Li, Shuo Fang, Zheng Faculty of Robot Science and Engineering Northeastern University Shenyang110819 China National Frontiers Science Center for Industrial Intelligence and Systems Optimization Northeastern University Shenyang110819 China Key Laboratory of Data Analytics and Optimization for Smart Industry Ministry of Education Northeastern University Shenyang110819 China

3D single object tracking with LiDAR points is an important task in the computer vision field. Previous methods usually adopt the matching-based or motion-centric paradigms to estimate the current target status. However, the former is sensitive to the similar distractors and the sparseness of point clouds due to relying on appearance matching, while the latter usually focuses on short-term motion clues (eg. two frames) and ignores the long-term motion pattern of target. To address these issues, we propose a mixed paradigm with two stages, named MTM-Tracker, which combines motion modeling with feature matching into a single network. Specifically, in the first stage, we exploit the continuous historical boxes as motion prior and propose an encoder-decoder structure to locate target coarsely. Then, in the second stage, we introduce a feature interaction module to extract motion-aware features from consecutive point clouds and match them to refine target movement as well as regress other target states. Extensive experiments validate that our paradigm achieves competitive performance on large-scale datasets (70.9% in KITTI and 51.70% in NuScenes). The code will be open soon at https://***/LeoZhiheng/ ***. Copyright © 2023, 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

A Deep Learning model for Question Analysis in Low-resource Languages: A Dataset and Case Study for Persian

A Deep Learning model for Question Analysis in Low-resource ...

引用

Pattern Recognition Systems (ICPRS), International Conference on

作者： Fatemeh Ebrahimi Khaksefidi Afsaneh Fatemi Mohammad Ali Nematbakhsh Mahsa Abazari Kia Faculty of Computer Engineering Department University of Isfahan Isfahan Iran Faculty of Computer and Data Science Northeastern University London London United Kingdom

ISBN: (数字)9798350375657

ISBN: (纸本)9798350375664

Question-answering systems, characterized by their fundamental functions of question classification, information retrieval, and answer selection, demand refinement to enhance precision in retrieving exact answers. Question classification, a cornerstone task, anticipates the probable answer to a posed query. However, the performance of question classification algorithms is hampered, particularly in agglutinative languages with complex morphology like Persian, where linguistic resources are limited. In this study, we propose a novel multi-layer Long-short-term memory (LSTM) Attention Convolutional Neural Network (CNN) (LACNN) classifier, tailored to extract pertinent information from Persian language contexts. Notably, this model operates autonomously, obviating the need for prior knowledge or external features. Moreover, we introduce UIMQC, the first medical question dataset in Persian, derived from the English GARD dataset. The inquiries within UIMQC are inherently intricate, often pertaining to rare diseases necessitating specialized diagnosis. Our experimental findings demonstrate a notable enhancement over baseline methods, with a 9% performance increase on the UTQC dataset, and achieving 67.08% accuracy on the UIMQC dataset. Consequently, we advocate for the adoption of the LACNN model in various morphological analysis tasks across low-resource languages, as in Question Answering systems it improves the performance for retrieving accurate answers to the users’ queries.

关键词： Analytical models Accuracy Refining Morphology Manuals Data augmentation Question answering (information retrieval)

来源：评论

学校读者我要写书评

暂无评论

Tackling Real-world Complexity: Hierarchical Modeling and Dynamic Prompting for Multimodal Long Document Classification

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年第6期35卷 5776-5790页

作者： Liu, Tengfei Hu, Yongli Li, Mingjie Yi, Junfei Chang, Xiaojun Gao, Junbin Yin, Baocai Beijing University of Technology Beijing Key Laboratory of Multimedia and Intelligent Software Technology Beijing Institute of Artificial Intelligence School of Information Science and Technology Beijing100124 China Stanford University Department of Radiation Oncology StanfordCA94305 United States Hunan University National Engineering Laboratory of Robot Visual Perception and Control Technology College of Electrical and Information Engineering Hunan Changsha410082 China University of Technology Sydney ReLER Laboratory Australian Artificial Intelligence Institute Faculty of Engineering and Information Technology UltimoNSW2007 Australia The University of Sydney Business School The University of Sydney Discipline of Business Analytics CamperdownNSW2006 Australia

With the rapid growth of internet content, multimodal long document data has become increasingly prominent, drawing significant attention from researchers. However, most existing methods primarily focus on scenarios where all modalities are present, often overlooking more challenging and realistic cases involving missing image modality. To address this limitation, we propose a robust multimodal long document classification (MLDC) framework that integrates hierarchical modeling and dynamic prompting to handle complex multimodal long document data. Our approach begins by leveraging hierarchical modeling combined with an Adaptive Correlation Multimodal Transformer (ACMT) to effectively capture relationships between text and images at both section and sentence levels. We also introduce a Dynamic Prompt Generation (DPG) module at both levels to enhance the model's robustness in handling missing image data. By evaluating sample uncertainty, the DPG module dynamically adjusts both the number of prompts and the prompts themselves, allowing the model to better adapt to the varying needs of different samples. Finally, a Hierarchical Heterogeneous Graph (HHG) is introduced to enhance feature interactions across levels, further improving the coherence and accuracy of the model. Extensive experiments on four multi-modal long document datasets demonstrate that our model shows superior performance compared to existing state-of-the-art MLDC classification methods in various conditions. © 1991-2012 IEEE.

关键词： Adaptation models Transformers Data models Correlation Circuits and systems Electronic mail Robustness Uncertainty Training Complexity theory

来源：评论

学校读者我要写书评

暂无评论

Surgical Assistance System Based on Visual Servoing

Surgical Assistance System Based on Visual Servoing

引用

第33届中国控制与决策会议

作者： Ruiqi Rao Fei Wang Sike Luan Qibin Chen Jiaqi Wei Faculty of Robot Science and Engineering Northeastern University College of Computer Science and Engineering Northeastern University

With the continuous improvement of medical services,surgical assisted robots have important application prospects,and the development of its technology has high requirements for the coordination of manipulator and doctor' s *** paper designs a surgical assistance system based on visual servoing to solve the problem of the manipulator dynamically following the doctor's *** system combines computer vision,automatic control principles,manipulator control and motion planning *** this paper,the AprilTag visual positioning algorithm is called to identify the AprilTag's posture,and the posture of the target point relative to the AprilTag is specified to determine the movement of the manipulator from the initial posture to the desired posture,and the Kalman filter algorithm is used to estimate the movement speed of AprilTag,carry out feedforward compensation to the servo system to correct the errors generated during the motion *** simulation experiments and physical experiments,the performance of the system is verified under different illumination conditions under the conditions of AprilTag static and *** experimental results show that the system has lower tracking error and higher *** can identify targets under different lighting conditions and different angles,and can dynamically follow the AprilTag on the doctor's hand to achieve the purpose of dynamically following the doctor' s *** has adaptive planning and anti-interference for both static and dynamic scenarios.

关键词： Visual servo Kalman filter Dynamic following Manipulator motion planning

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：