检索结果-内蒙古大学图书馆

6th IEEE Advanced Information Management, Communicates, Electronic and Automation Control Conference, IMCEC 2024

作者： Xiaoting, Dong Guangxi, Wan Peng, Zeng Shenyang Institute of Automation Chinese Academy of Sciences Key Laboratory of Networked Control Systems Shenyang110016 China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang110169 China University of Chinese Academy of Sciences Beijing100049 China State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Shenyang110016 China

ISBN: (纸本)9798350316520

In flexible job shop, the transportation time and path of Automated Guided Vehicles (AGVs) greatly impact the production cycle time. However, traditional flexible job shop scheduling problems (FJSSP) mainly focus on job sequencing and machine selection, neglecting the effect of AGVs transportation on the production process. This often leads to suboptimal scheduling solutions that are difficult to guide the shop floor operation effectively. To address this issue, this paper formulates a cooperative scheduling model that integrates the AGVs scheduling into traditional flexible job shop scheduling problem, abbreviated as the FJSMAS problem, with the objective of makespan minimization. Furthermore, by designing state features, action space, and reward function, the FJSMAS problem is transformed into a Markov decision process, a deep reinforcement learning algorithm (DRL) is then adopted to solve it. The feasibility of the proposed FJSMAS model and the effectiveness of DRL algorithm in solving the FJSMAS problem are verified through benchmark's instances. © 2024 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

GCA-Net: A Global Context Aggregation Network for Effective Optical Flow

GCA-Net: A Global Context Aggregation Network for Effective ...

引用

Annual Conference of Industrial Electronics Society

作者： Tao Xie Jinghan Gao Ke Wang Ruifeng Li State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin China

Optical flow seeks to estimate the per-pixel 2D motion between two frames by identifying corresponding pixels. Current flow estimators typically involve per-pixel feature extraction, multi-scale 4D correlation volume construction, and iterative flow field updates through a Conv-GRU module. However, the locality of convolutional features in these methods renders the calculated correlations vulnerable to different noises. In addition, the Conv-GRU module of these methods is only executed with a convolution layer, which is incapable of exploiting context clues from larger window sizes even inside the image being queried itself, thus raising it more difficult for the model to process images with challenging regions, e.g., textureless areas. To this end, we introduce GCA-Net, a global context aggregation network for credible yet effective optical flow estimation. More specifically, we propose a highly efficient multi-scale transformer (MSFormer) layer which enables the per-pixel feature to aggregate long-range information from other features, hence building more accurate 4D correlation volumes. Besides, we develop an attention-enhanced Conv-GRU block (AttGRU) that can incorporate information alongside a larger context window even within itself empowering our network to estimate optical flow in even the most challenging regions. Experimentally, we demonstrate that GCA-Net outperforms previous state-of-the-art methods by large margins on Sintel (Final) and KITTI 2015 (background and foreground) benchmarks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Few-Shot Object Detection via Dual-Domain Feature Fusion and Patch-Level Attention

引用

清华大学学报自然科学版（英文版） 2025年第3期30卷 1237-1250页

作者： Guangli Ren Jierui Liu Mengyao Wang Peiyu Guan Zhiqiang Cao Junzhi Yu State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of AutomationChinese Academy of SciencesBeijing 100190China School of Artificial Intelligence University of Chinese Academy of SciencesBeijing 100049China Department of Advanced Manufacturing and Robotics Peking UniversityBeijing 100871China

Few-shot object detection receives much attention with the ability to detect novel class objects using limited annotated *** transfer learning-based solution becomes popular due to its simple training with good accuracy,however,it is still challenging to enrich the feature diversity during the training *** fine-grained features are also insufficient for novel class *** deal with the problems,this paper proposes a novel few-shot object detection method based on dual-domain feature fusion and patch-level *** original base domain,an elementary domain with more category-agnostic features is superposed to construct a two-stream backbone,which benefits to enrich the feature *** better integrate various features,a dual-domain feature fusion is designed,where the feature pairs with the same size are complementarily fused to extract more discriminative ***,a patch-wise feature refinement termed as patch-level attention is presented to mine internal relations among the patches,which enhances the adaptability to novel *** addition,a weighted classification loss is given to assist the fine-tuning of the classifier by combining extra features from FPN of the base training *** this way,the few-shot detection quality to novel class objects is *** on PASCAL VOC and MS COCO datasets verify the effectiveness of the method.

关键词： few-shot object detection dual-domain feature fusion patch-level attention

来源：评论

学校读者我要写书评

暂无评论

Lower limb motion pattern recognition based on IWOA-SVM 3

Lower limb motion pattern recognition based on IWOA-SVM

引用

3rd International Conference on Computer Science and Communication Technology, ICCSCT 2022

作者： Li, Xiaoqi Yang, Yunlong Chen, Haidie Yao, Yufeng State Key Laboratory of Robotics and Systems Harbin Institute of Technology Heilongjiang Harbin China

ISBN: (纸本)9781510661240

This paper presents a lower limb motion pattern recognition algorithm based on SVM model optimized by improved whale algorithm (IWOA-SVM). The purpose is to overcome the problems of low accuracy and low reaction speed of existing lower limb rehabilitation pattern recognition algorithms, nonlinear adjustment factor is combined with WOA to enhance the search ability and search speed of the algorithm. The collected surface EMG signals are used as the input of the motion pattern recognition system, and the motion pattern recognition is realized by combining the IWOA-SVM model. Simulation experiments on six test functions are carried out and compared with WOA. The results indicated that the search convergence speed and optimization accuracy of IWOA are improved. Experiments on the collected signal data suggested that the recognition accuracy is 94.12%, compared with WOA-SVM algorithm, PSO-SVM and GA-SVM. The results indicated that the recognition accuracy of this algorithm is improved by 4.02%, 8.82% and 6.64% respectively. © 2022 SPIE.

关键词： Pattern recognition

来源：评论

学校读者我要写书评

暂无评论

Usage of Colloidal AFM Probe for Research in Effects of Water Layer Evaporation on Interfacial Adhesion between Cellulose Surfaces 12

Usage of Colloidal AFM Probe for Research in Effects of Wate...

引用

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

作者： Geng, Junyuan Zhang, Hao Meng, Xianghe Gao, Haibo Rong, Weibin Xie, Hui Harbin Institute of Technology State Key Laboratory of Robotics and Systems Harbin150080 China

ISBN: (数字)9781665472678

ISBN: (纸本)9781665472678

Effects of the water layer evaporation on the interfacial adhesion between cellulose surfaces are researched via the atomic force microscopy (AFM) method in this article. A cellulose colloidal probe (CCP) is prepared through a dual-probe micromanipulation system, which is used to detect this time-varying adhesive interaction. The AFM system configuration, detailed fabrication of the CCP, principle of the adhesion measurement, and relevant calibration tasks are reported before the characterization. Several groups of adhesion mapping experiments have been executed at different divided periods after the water film forming on the sample surface of the cellulose membrane. The obtained results show that the interfacial adhesion has a significant downward trend when the water layer passing over time. In addition, the correlation of the surface topography and adhesion is displayed. The industrial applications of cellulose are inseparable from the participation of water solvent, meanwhile appropriate process adjustments according to the bonding strength between the cellulose surfaces during pulping and drying is necessary. This research can provide support in above occasions, which has its unique significance on the improving of processing technology. © 2022 IEEE.

关键词： Cellulose

来源：评论

学校读者我要写书评

暂无评论

A Practical Multi-camera SLAM system for Large Mobile Robots 2

A Practical Multi-camera SLAM System for Large Mobile Robots

引用

2nd International Conference on Big Data, Artificial Intelligence and Risk Management, ICBAR 2022

作者： Dong, Yunlong Ding, Hong Zha, Fusheng Li, Mantian State Key Laboratory of Robotics and Systems Harbin Institute of Technology Harbin150080 China

ISBN: (纸本)9798350334265

It is very important to provide sufficient visual perception for large mobile robots to improve their security and intelligence. Considering the narrow field of vision and insufficient perception of a single camera, this paper proposes a practical multi-camera SLAM system. According to the hardware structure of the large mobile robot, the layout scheme of the multi-camera is designed and its field of view is analyzed. To make full use of the advantages of multi-camera data fusion, a multi-camera localization algorithm based on the main localization module and sub-localization module is proposed. In order to build a dense 3D point cloud map, a novel depth estimation method is present, which first uses the depth estimation method based on camera geometry to obtain a preliminary depth map, and then uses a convolutional neural network to refine it. Comprehensive experiments prove that our multi-camera SLAM system achieves appealing results and has strong practicability. © 2022 IEEE.

关键词： Mobile robots

来源：评论

学校读者我要写书评

暂无评论

Cam-Driven Frequency Up-Conversion Mechanism For Kinetic Energy Harvesting

Cam-Driven Frequency Up-Conversion Mechanism For Kinetic Ene...

引用

Micro and Nanotechnology for Power Generation and Energy Conversion Applications (PowerMEMS)

作者： Heng Zhao Tianyi Tang Yunfei Li Mingqi Mei Huicong Liu School of Mechanical and Electrical Engineering Jiangsu Provincial Key Laboratory of Advanced Robotics Soochow University Suzhou CHINA State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin CHINA

ISBN: (数字)9798350380200

ISBN: (纸本)9798350380217

This paper proposes an ocean wave energy harvester (WEH) with a cam-driven frequency up-conversion mechanism. It utilizes a planar pendulum structure to achieve motion response to ultra-low-frequency wave excitation and then converts the mechanical energy of the pendulum into high-frequency and high-power electrical energy output through the cam-driven frequency upconversion mechanism. WEH’s superior design enables high-power output even under excitation below 1 Hz. At the excitation frequency of 0.75 Hz and resistance load of $20 \Omega$, the maximum power output can reach 461.4 mW. Due to its high output power, the WEH can be integrated into the ocean observation buoy to power the sensors, thereby enabling long-term, unattended operation of the buoy in remote offshore areas.

关键词： Temperature sensors Resistance Ocean waves Frequency conversion Sensor systems Sensors Power systems Wave energy conversion Power generation Ocean temperature

来源：评论

学校读者我要写书评

暂无评论

Poly-MOT: A Polyhedral Framework For 3D Multi-Object Tracking

Poly-MOT: A Polyhedral Framework For 3D Multi-Object Trackin...

引用

IEEE/RSJ International Conference on Intelligent Robots and systems (IROS)

作者： Xiaoyu Li Tao Xie Dedong Liu Jinghan Gao Kun Dai Zhiqiang Jiang Lijun Zhao Ke Wang State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin China

3D Multi-object tracking (MOT) empowers mobile robots to accomplish well-informed motion planning and navigation tasks by providing motion trajectories of surrounding objects. However, existing 3D MOT methods typically employ a single similarity metric and physical model to perform data association and state estimation for all objects. With large-scale modern datasets and real scenes, there are a variety of object categories that commonly exhibit distinctive geometric properties and motion patterns. In this way, such distinctions would enable various object categories to behave differently under the same standard, resulting in erroneous matches between trajectories and detections, and jeopardizing the reliability of downstream tasks (navigation, etc.). Towards this end, we propose Poly-MOT, an efficient 3D MOT method based on the Tracking-By-Detection framework that enables the tracker to choose the most appropriate tracking criteria for each object category. Specifically, Poly-MOT leverages different motion models for various object categories to characterize distinct types of motion accurately. We also introduce the constraint of the rigid structure of objects into a specific motion model to accurately describe the highly nonlinear motion of the object. Additionally, we introduce a two-stage data association strategy to ensure that objects can find the optimal similarity metric from three custom metrics for their categories and reduce missing matches. On the NuScenes dataset, our proposed method achieves state-of-the-art performance with 75.4% AMOTA. The code is available at https://***/lixiaoyu20001P0Iy-MOT.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Trajectory optimization of B-splines interpolation based on dynamic error adjustment 9

Trajectory optimization of B-splines interpolation based on ...

引用

2023 9th International Forum on Manufacturing Technology and Engineering Materials, ICMTEM 2023

作者： Zhao, Peng Gao, Feiyang Guo, Kai Zhang, Enyuan Li, Song School of Mechanical Engineering Shenyang Ligong University Liaoning Shenyang110159 China State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Liaoning Shenyang110016 China Institute of Robotics and Intelligent Manufacturing Innovation Chinese Academy of Sciences Liaoning Shenyang110169 China

In this paper, we propose a motion planning algorithm based on the five-fold B-spline curve for multi-axis motion control systems with non-smooth trajectories, acceleration, plus acceleration over the limit, etc. We construct a smooth transition model for the micro-line segment G3 of the continuous five-fold B-spline curve. A derivation of the corresponding theoretical equation is given. Using the transition model, we propose an optimization method for dynamic adjustment of motion trajectory errors based on the velocity forward planning constraints by smoothing the transition after the trajectory of the micro-line segment for the forward planning velocity. Finally, the design flow for the forward planning of the velocity is given. Finally, the above algorithm is used to perform simulation experiments for the optimization algorithm. Experimental results show that the proposed optimization improves the accuracy of fitting motion trajectories under the constraints of acceleration and deceleration addition, which validates the proposed algorithm. © Published under licence by IOP Publishing Ltd.

关键词： Acceleration

来源：评论

学校读者我要写书评

暂无评论

Cable Length Estimation for a Hyper-Redundant Robot Based on Cable Hole Clearance Calculation 19

Cable Length Estimation for a Hyper-Redundant Robot Based on...

引用

19th IEEE International Conference on Automation Science and Engineering, CASE 2023

作者： Li, Ziqing Deng, Jiangqin Zheng, Yang Liu, Chao Gu, Guoying Robotics Institute School of Mechanical Engineering Shanghai Jiao Tong University Shanghai200240 China Shanghai Jiao Tong University State Key Laboratory of Mechanical System and Vibration Shanghai200240 China

ISBN: (纸本)9798350320695

For the cable-driven hyper-redundant robot, most of the current drive cable length estimation methods consider that the cable always passes through the center of the hole, and do not consider the effect of cable-hole clearance, resulting in a large motion error. In this paper, we propose a cable length estimating method by considering cable-hole clearance. Based on the simplified space model of the cable hole, it uses the recursive algorithm to obtain the approximate solution of the actual drive cable length. The algorithm can meet the requirements of real-time control while ensuring calculation accuracy. The correctness of the approximate solution is verified by using the interior point penalty function method. Taking a 24DoF hyper-redundant manipulator with joint angle closed-loop control as an example, the effectiveness of the proposed method is proved by comparing the theoretical calculation results with the actual drive cable length. © 2023 IEEE.

关键词： Real time control

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：