检索结果-内蒙古大学图书馆

3rd International Conference on Artificial Intelligence and Education, ICAIE 2024

作者： Wang, Xiao Chu, Hao Ji, Peng Yu, Xiaosheng College Of Information Science and Engineering Northeastern University Liaoning Shenyang China Faculty of Robot Science and Engineering Northeastern University Liaoning Shenyang China

ISBN: (纸本)9798400712692

Traditional AI-related courses in robotics major are not related closely to the latest technological advancements. In order to explore a better approach in the teaching system, this paper introduces a new AI course cluster system. The system conducts the PBL teaching structure and designs a tailored AI course cluster for robotics students. An organized and project-based teaching model is established too. We also conduct case studies to the students in the robotics major. The experimental results verify the effective of the proposed cluster system. © 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

SGT-Net: A Transformer-Based Stratified Graph Convolutional Network for 3D Point Cloud Semantic Segmentation

引用

Computers, Materials & Continua 2024年第6期79卷 4471-4489页

作者： Suyi Liu Jianning Chi Chengdong Wu Fang Xu Xiaosheng Yu Faculty of Robot Science and Engineering Northeastern UniversityShenyang110167China State Key Laboratory of Robotics Shenyang Institute of AutomationChinese Academy of SciencesShenyang110016China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of SciencesShenyang110169China SIASUN Robot&Automation Co. Ltd.Shenyang110169China

In recent years,semantic segmentation on 3D point cloud data has attracted much *** 2D images where pixels distribute regularly in the image domain,3D point clouds in non-Euclidean space are irregular and inherently ***,it is very difficult to extract long-range contexts and effectively aggregate local features for semantic segmentation in 3D point cloud *** current methods either focus on local feature aggregation or long-range context dependency,but fail to directly establish a global-local feature extractor to complete the point cloud semantic segmentation *** this paper,we propose a Transformer-based stratified graph convolutional network(SGT-Net),which enlarges the effective receptive field and builds direct long-range ***,we first propose a novel dense-sparse sampling strategy that provides dense local vertices and sparse long-distance vertices for subsequent graph convolutional network(GCN).Secondly,we propose a multi-key self-attention mechanism based on the Transformer to further weight augmentation for crucial neighboring relationships and enlarge the effective receptive *** addition,to further improve the efficiency of the network,we propose a similarity measurement module to determine whether the neighborhood near the center point is *** demonstrate the validity and superiority of our method on the S3DIS and ShapeNet *** ablation experiments and segmentation visualization,we verify that the SGT model can improve the performance of the point cloud semantic segmentation.

关键词： 3D point cloud semantic segmentation long-range contexts global-local feature graph convolutional network dense-sparse sampling strategy

来源：评论

学校读者我要写书评

暂无评论

Finite-Time Event-Triggered Containment Maneuvering of Marine Surface Vehicles With Tracking Error Constraints: Theory and Experiment

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on Intelligent Vehicles 2024年 1-11页

作者： Feng, Yuxi Wang, Hao Fu, Jun Faculty of Robot Science and Engineering Northeastern University Shenyang China

In this article, a novel class of containment maneuvering controllers is developed for a fleet of marine surface vehicles (MSVs) capable of tracking a parameterized path within finite time. Besides, the along-track errors and cross-track errors of parameterized line-of-sight (LOS) are constrained. For kinematic level, to address tracking error constraints and finite time convergence, the coordinated guidance laws are derived by developing LOS based tan-type barrier Lyapunov functions (BLFs), nonlinear tracking differentiator (NLTD) and finite-time theory. Specifically, a novel finite-time extended state observer (FTESO) is developed to estimate the uncertain kinematics. For coordination level, the finite-time switching threshold event-triggered containment control strategy for MSV is derived to reduce the updating frequency of the path variables containment controller. Theoretical analysis illustrates that all signals are ultimately uniformly bounded (UUB). Both simulation and experimental results are provided to verify the effective performance of the proposed containment maneuvering strategy. IEEE

关键词： Kinematics

来源：评论

学校读者我要写书评

暂无评论

HyGFNet: Hybrid Geometry-Flow Learning Network for 3D Single Object Tracking

IEEE Transactions on Intelligent Vehicles

引用

IEEE Transactions on Intelligent Vehicles 2024年 1-12页

作者： Cui, Yubo Fang, Zheng Li, Zhiheng Li, Shuo Lin, Yu Faculty of Robot Science and Engineering Northeastern University Shenyang China

3D single object tracking (SOT) which attempts to accurately locate the target object in the current frame, has made significant advancements over past years. However, most previous works built upon the Siamese architecture usually focus on the learning and matching of geometry information, while neglecting the motion information of the target. Consequently, those methods face challenges when distinguishing the target object from similar distractors. To overcome this limitation, we enhance the architecture by incorporating flow estimation, presenting a novel multi-frame hybrid geometry-flow learning network for 3D SOT. The proposed framework exploits both geometry and flow information from historical frames and further integrates the learned information into the current frame to improve the target object localization. Specifically, we propose a geometry matching branch and a flow estimation branch. The geometry matching branch first captures the geometry feature of each frame and then aligns these features through multi-frame spatial-temporal matching, leading to a geometry-aware feature for the target object. Meanwhile, in the flow estimation branch, a multi-frame flow-aware enhancement module is proposed to explicitly capture flow information across frames, leading to a flow-aware feature for the target object. Finally, the geometry-aware and flow-aware features are fused with the original feature to predict the position of the target object. Extensive experiments conducted on the KITTI and nuScenes datasets validate the effectiveness and show the competitive performance of our method. The code will be open soon. IEEE

关键词： Geometry

来源：评论

学校读者我要写书评

暂无评论

Salient Object Detection via A Novel Saliency Ranking Tree Integrating Low-level and High-level Features 15

Salient Object Detection via A Novel Saliency Ranking Tree I...

引用

15th International Conference on Graphics and Image Processing, ICGIP 2023

作者： Pang, Yu Huang, Yang Yu, Xiaosheng Wu, Chengdong School of Artificial Intelligence Shenyang University of Technology Shenyang110870 China Faculty of Robot Science and Engineering Northeastern University Shenyang110819 China

ISBN: (纸本)9781510675018

In this paper, we propose a novel salient object detection framework by constructing a novel saliency tree model integrating low-level and high-level features. In our model, numerous features containing low-level features (e.g., color, texture, gradient, contrast, etc.) and high-level features (e.g., deep features extracted from pre-trained VGG19 net) are firstly selected as candidate features. We develop a novel feature integrating mechanism to acquire an integrated feature descriptor which is more discriminative to capture the contrast between foreground and background for the input image. Then, we construct a novel saliency tree model relied on the integrated features to generate saliency map. We compare the proposed method and other state-of-the-art methods on three datasets, experimental results indicate that the proposed saliency detection algorithm has achieved the top performance. © 2024 SPIE. All rights reserved.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

Semantic Range Image for Loop Closure of 3D Lidar SLAM 13

Semantic Range Image for Loop Closure of 3D Lidar SLAM

引用

13th IEEE International Conference on CYBER Technology in Automation, Control, and Intelligent Systems, CYBER 2023

作者： Li, Fang Shen, Yehui Zou, Fengshan Du, Zhenjun Liu, Mingmin Northeastern University Faculty of Robot Science and Engineering Liaoning Shenyang110169 China SIASUN Robot & Automation Co. Ltd Liaoning Shenyang110168 China

ISBN: (纸本)9798350315196

The purpose of SLAM loop closure detection is to reduce the cumulative error in the system by identifying the same scene that the robot passed through. Traditional loop detection methods detect loops based on physical information. These methods use basic physical features to identify loop closures and pay less attention to the role of semantic information as high-level information in loop closure detection. This paper regards semantic information as one of the criteria for confirming loop closure, so that semantic information can be deeply involved in the process of loop closure detection. We construct a global descriptor based on range image and semantic information, which is called SeRI. The descriptor encodes the semantic and geometric features of the surrounding environment. Loop closure candidates are determined by matching between pairs of semantic range images. We evaluate our method on the KITTI dataset. Experiments have shown that the proposed method can effectively identify loop closure candidates correctly with high precision and recall rate. © 2023 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Trajectory Planning of robot End-Effector Position and Orientation Based on Cubic Spline 5

Trajectory Planning of Robot End-Effector Position and Orien...

引用

5th World robot Conference Symposium on Advanced robotics and Automation, WRC SARA 2023

作者： Liu, Yang Du, Zhenjun Liu, Mingmin Tao, Yongliang Li, Chenpeng Northeastern University Faculty of Robot Science and Engineering Liaoning Shenyang110169 China Siasun Robot & Automation Co. Ltd Liaoning Shenyang110168 China

ISBN: (纸本)9798350307320

Trajectory planning for the end effector of industrial robots has been a hot topic in the field of robotics. A continuous and smooth trajectory of the end effector's pose is beneficial for avoiding unnecessary impacts and vibrations, thereby improving the performance and lifespan of industrial robots. In recent years, researchers have predominantly focused on the trajectory planning of multi-directional variations in orientation using quaternions, while giving less attention to the coordinated planning of position and orientation. To address this issue, this paper proposes an efficient trajectory planning method based on improved spline curve interpolation. The orientation of key pose points, characterized by multi-directional variations, is represented using quaternions. By employing a quaternion dimension reduction algorithm, these quaternions are transformed into vector forms in three-dimensional space, allowing for unified representation of position and orientation within the same dimension. Taking the chord length between key pose points as a parameter, the position and orientation components are subjected to cubic spline curve interpolation with chord length as the independent variable. The proposed method is applicable for coordinated planning of position and orientation trajectories, ensuring high-order continuity of the pose trajectory. The effectiveness and rationality of the proposed method are validated through a set of key pose points featuring multi-directional variations. Finally, a comparative analysis is conducted between this set of key pose points and existing trajectory planning algorithms. Experimental results demonstrate that the proposed method not only guarantees high-order continuity of the trajectory but also exhibits higher smoothness. © 2023 IEEE.

关键词： Trajectories

来源：评论

学校读者我要写书评

暂无评论

Enhancing Incremental Learning with One-Class Envelopes for Online Unlabeled Mixed Samples 16

Enhancing Incremental Learning with One-Class Envelopes for ...

引用

16th International Conference on Digital Image Processing, ICDIP 2024

作者： Hu, Chengyong Ding, Qichuan Faculty of Robot Science and Engineering Northeastern University Shenyang110169 China Foshan Graduate School of Innovation Northeastern University Foshan528311 China

ISBN: (数字)9781510682917

ISBN: (纸本)9781510682900

Incremental learning is crucial for handling classification tasks involving continuously increasing new categories. Traditional models often grapple with challenges such as data forgetting and high storage requirements. An approach grounded in one-class envelope techniques effectively mitigates these concerns. However, this method introduces two new challenges: swiftly determining the class label of a new sample, and effectively distinguishing mixed samples online before training new sub-classifiers. To address these issues, this paper introduces an efficient incremental learning strategy that leverages one-class envelopes, where a dedicated one-class classifier is built for each category. A novel hybrid-distance-based search algorithm for the target category is introduced to enhance prediction efficiency during the testing phase. Moreover, for online mixed samples from multiple unknown categories, the K-means algorithm is employed to cluster all outlier samples. Silhouette coefficients are then calculated to identify well-defined clusters for subsequent classifier training. Experimental results show that the proposed approach reduce prediction time by 90% on Cifar100-30 dataset, and the incremental process for unlabeled mixed samples on the Cifar10 dataset achieves up to 55.3% performance improvement. © 2024 SPIE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

An Attention Based Chinese Sign Language Recognition Method Using sEMG Signal 12

An Attention Based Chinese Sign Language Recognition Method ...

引用

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

作者： Zeng, Zhen Wang, Fei Faculty of Robot Science and Engineering Northeastern University Shenyang110819 China

ISBN: (数字)9781665472678

ISBN: (纸本)9781665472678

This paper proposes a Chinese sign language recognition method based on attention mechanism. We use multimodal information from sEMG, ACC and GYR channels to explore the problem of Chinese sign language isolated word recognition based on EMG armbands. Our network adopts a CNN-RNN architecture to extract features of the data itself and time-domain features. the attention mechanism is also introduced into our method, and we use SE-Block to set the corresponding weights for different channels of the data. In the experimental part, we recruited 10 volunteers to collect a total of 50 kinds of 2780 data for us. We validated the effectiveness of our model with the dataset and achieved an overall accuracy of 89% and an F1 score of 86%. in terms of timeliness, our model analyzes individual sign languages in about 0.01 seconds, which is computationally low. Finally, we analyze the application scenarios and feasibility of this method. © 2022 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Solution method of coupling joint angle of discrete-continuous manipulator 7

Solution method of coupling joint angle of discrete-continuo...

引用

7th International Conference on Intelligent Computing and Signal Processing, ICSP 2022

作者： Fang, Lijin Zhou, Zuyang Lu, Yunhua Xu, Jiqian Faculty of Robot Science and Engineering Northeastern University Shen Yang China

ISBN: (纸本)9781665478571

Recently, continuum robots capable of probing in narrow spaces have attracted a lot of attention from researchers. However, it is difficult for the current continuum robot to quickly calculate the rotation angle of each joint, so it is hard to achieve precise control of the end position. This paper design a discrete-continuous manipulator structure driven by a double tendon sheath. Based on the mechanical model of the manipulator's joint disk, establish the calculation model of the coupling angle between adjacent joints and give the relationship between the input force and the end joint angle. The method for solving the angle of each joint of the discrete-continuous manipulator is given. This model solves the problem that the angle of each joint of the discrete-continuous manipulator is hard to solve quickly, and provides a basis for the subsequent precise control of the end position of the manipulator. © 2022 IEEE.

关键词： Manipulators

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：