检索结果-内蒙古大学图书馆

2nd International Conference on Computer Vision and Data Mining, ICVDM 2021

作者： Yang, Yunxiao Wangjun Zhang, Hualiang Dai, Shilong School of Computer Science Shenyang University of Chemical Technology Shenyang110142 China Industrial Control Network and System Laboratory Shenyang Institute of Automation Chinese Academy of Sciences Shenyang110016 China Institudes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang110169 China

Aiming at the problem of overestimation and sparse rewards of deep Q network algorithm in mobile robot path planning in reinforcement learning, an improved algorithm HERDDQN is proposed. Through the deep convolutional neural network model, the original RGB image is used as input, and it is trained through an end-to-end method. The improved deep reinforcement learning algorithm and the deep Q network algorithm are simulated in the same two-dimensional environment. The experimental results show that the HERDDQN algorithm solves the problem of overestimation and sparse reward better than the DQN algorithm in terms of success rate and reward convergence speed, Which shows that the improved algorithm finds a better strategy than the DQN algorithm. © 2021 Institute of Physics Publishing. All rights reserved.

关键词： Mobile robots

来源：评论

学校读者我要写书评

暂无评论

Point Cloud based Motion State Estimation Method for Autonomous Driving

Point Cloud based Motion State Estimation Method for Autonom...

引用

International Conference on Advanced Mechatronic systems (ICAMechS)

作者： Tao Jiang Enzhi Zhu Sidong Wu Linshuai Zhang Shuoxin Gu Lin Xu Qian Wang Xianggui Tang School of Intelligent Medicine Chengdu University of Traditional Chinese Medicine Unmanned System Intelligent Perception Control Technology Engineering Laboratory of Sichuan Province Chengdu University of Information Technology Chengdu School of Automation Chengdu University of Information Technology Chengdu School of Automation Chengdu University of Information Technology Unmanned System Intelligent Perception Control Technology Engineering Laboratory of Sichuan Province Chengdu University of Information Technology Chengdu School of Intelligent Medicine Chengdu University of Traditional Chinese Medicine School of Automation Chengdu University of Information Technology International Joint Research Center of Robotics and Intelligence System of Sichuan Province Chengdu School of Automation Chengdu University of Information Technology International Joint Research Center of Robotics and Intelligence System of Sichuan Province Chengdu School of Intelligent Medicine Chengdu University of Traditional Chinese Medicine Chengdu

Generally, motion vector information primarily stems from moving objects, while static objects contribute minimally to the estimation task. Traditional approaches for motion vector estimation typically rely on scene flow methods that depend on deep models to extract features from individual points at a high cost. These methods then acquire flow information through complex matching mechanisms or feature decoding. Such approaches are computationally expensive and exhibit substantial latency. Moreover, they neglect the importance of motion objects in motion vector estimation and the interference from static objects. Therefore, this paper introduces a novel method that first performs point cloud motion segmentation and subsequently estimates motion vectors. This approach focuses on leveraging point cloud information annotated with motion objects to estimate three-dimensional scene flow more effectively. By employing motion segmentation, we can obtain annotations for moving objects, enabling greater emphasis on the estimation of motion vectors for more challenging cases. In our experiments conducted on the KITTI dataset, the proposed method demonstrates superior performance compared to existing scene flow estimation methods. Specifically, without considering motion segmentation errors, the error in the motion direction is only 0.0363 m/s, showcasing better performance. Additionally, our method achieves an error of 0.076 m for three-dimensional endpoint error (EPE3D), showcasing distinct advantages over current scene flow networks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Miniaturized Pneumatic Electrode Gripper for Robotic Cochlear Implant Surgery

A Miniaturized Pneumatic Electrode Gripper for Robotic Cochl...

引用

IEEE International Conference on robotics and Biomimetics

作者： Hongyan Gao Yun Zou Shaoping Huang Weidong Chen Anzhu Gao Institute of Medical Robotics and Department of Automation Key Laboratory of System Control and Information Processing Ministry of Education Shanghai Jiao Tong University Shanghai P. R. China Institute of Medical Robotics Shanghai Jiao Tong University Shanghai P. R. China

ISBN: (纸本)9781665481106

Cochlear implant surgery is one of the most challenging surgeries due to the limited workspace and delicate inner structure. A traditional electrode implant instrument is a passive elastic forceps, which needs another forceps to help grip or release the electrode. It is not convenient to achieve delicate operations in a confined space. This paper develops a miniaturized pneumatic electrode gripper at the tip of forceps to achieve active gripping. It transmits the pressurized air through 3D-printed inner lumens to form two inflated soft bubbles to push the electrode to the rigid shell for accurate positioning. The finite element analysis is carried out to evaluate the shape of inflated soft bubbles with variable soft films. A prototype of a gripper with inner tortuous airways is fabricated using micro 3D printing technique, and experiments are carried out to validate its accuracy and demonstrate its ability to grip and release. Finally, robotic teleoperation using a robotic arm and a joystick is implemented for cochlear implant surgery. Results demonstrate that the developed miniaturized pneumatic gripper has successfully achieved electrode implantation in a 3D-printed cochlear phantom.

关键词： Electrodes Cochlear implants Shape Surgery Prototypes Phantoms Pneumatic systems

来源：评论

学校读者我要写书评

暂无评论

Efficient Planning for Object Search Task Based on Hierarchical POMDP

Efficient Planning for Object Search Task Based on Hierarchi...

引用

IEEE International Conference on robotics and Biomimetics

作者： Wenrui Zhao Weidong Chen Institute of Medical Robotics Shanghai Jiao Tong University and Key Laboratory of System Control and Information Processing Ministry of Education Shanghai China

ISBN: (纸本)9781665405362

Searching for a target object in clutter by a robot needs to solve perception uncertainty due to occlusion among objects, the robot has to move objects to gather information about where the target may locate. Meanwhile, we hope that the robot could execute as few actions as possible to get the target. Partially observable Markov decision process (POMDP) is proper to model this task, but it is hard to solve for large scale problems. And action feasibility of robotic manipulation must be identified by motion planning, which is time-consuming. We adopt hierarchical POMDP to reduce the complexity of POMDP planning. Besides, we propose methods to reduce the times of motion planning, the key idea is assigning reward to actions related to feasibility probability, instead of identifying action feasibility at the beginning. Simulations under various scenes show that our methods are effective and efficient.

关键词： Uncertainty Conferences Biomimetics Markov processes Search problems Planning Complexity theory

来源：评论

学校读者我要写书评

暂无评论

Study on Mechanical Properties of Dual-Channel Cryogenic 3d Printing Scaffold for Mandibular Defect Repair

SSRN

引用

SSRN 2023年

作者： Gao, Lilan Sun, Mengchao Liu, Jie Meng, Lulu Liu, Han Zheng, Xiongfei Liu, Hao Li, Ruixin Tianjin Key Laboratory for Advanced Mechatronic System Design and Intelligent Control School of Mechanical Engineering Tianjin University of Technology Tianjin300382 China Tianjin300382 China Beijing Friendship Hospital Capital Medical University 100050 China State Key Laboratory of Robotics Institutes for Robotics and Intelligent Manufacturing Shenyang Institute of Automation Chinese Academy of Sciences Shenyang110017 China Tianjin Key Laboratory of Oral and Maxillofacial Function Reconstruction Tianjin Stomatological Hospital The Affiliated Stomatological Hospital of Nankai University Tianjin300041 China

Mandibular defect repair has always been a clinical challenge, facing technical bottleneck. The new materials directly affect technological breakthroughs in mandibular defect repair field. Our aim is to fabricate a scaffold for mandibular defect repairing with advanced biomaterials. Therefore, a novel dual-channel printed scaffold consisting of silk fibroin/collagen type I /hydroxyapatite (SCH) and polycaprolactone/hydroxyapatite (PCL/HA) was fabricated by 3D printing technology with double nozzles. The mechanical properties and behaviors of dual-channel printed scaffold were investigated by performing uniaxial compression, creep, stress relaxation and ratcheting experiments respectively. The experiments indicated the Young modulus of dual-channel printed scaffold was 60.1 kPa and dual-channel printed scaffold was typical nolinear viscoelastic consistent with bone tissue. Finite Element Analysis (FEA) was employed performing a numerical simulation to evaluate the implantation effect in mandible. The stress distribution of the contact area between scaffold and defect was uniform, the maximum Mises stress of cortical bone and cancellous bone in defect area were 54.52MPa and 3.196MPa, and the maximum displacement of cortical bone and cancellous bone in defect area were 0.1575mm and 0.1555mm, which were distributed in the incisor region. The peak value of the maximum Mises stress for the scaffold implanted was 3.128× MPa and the maximum displacement was 6.453×mm distributed near incisor area. The displacement distribution of the scaffold was consistent with that of cortical and cancellous bone. The scaffold recovered well when the force applied on it disappeared. Above all, dual-channel printed scaffold had excellent bio-mechanical properties in implanting mandible, which provides a new idea for the reconstruction of irregular bone defects in the mandible and has good clinical development prospects. © 2023, The Authors. All rights reserved.

关键词： 3D printing

来源：评论

学校读者我要写书评

暂无评论

Analysis of electricity loss and electricity consumption law in low-voltage areas: A case study 2

Analysis of electricity loss and electricity consumption law...

引用

2nd International Conference on Advanced Electrical and Energy systems, AEES 2021

作者： Chen, Xi Song, Chunhe Wang, Tianran Key Laboratory of Network Control System Chinese Academy of Sciences Shenyang110016 China Shenyang Institute of Automation Chinese Academy of Sciences Shenyang110016 China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang110016 China University of Chinese Academy of Sciences Beijing100049 China

Electric power is an important resource in energy. With the development of society, the consumption of electric power is more and more. Research on electricity consumption and power loss has become a hot *** this paper, we conduct statistical analysis on the power supply, power consumption and line loss rate of 395 power supply *** loss rate is an indicator to describe electricity *** found that the line loss rate decreased over the course of a *** overall distribution of line loss rate is approximately lognormal distribution. Total daily electricity supply and electricity consumption in the power supply area approximately obey exponential *** terms of electricity consumption, a statistical analysis was made on the daily electricity consumption of 100,000 users in three years, and it was found that the electricity distribution of users was in line with the power law distribution. © 2021 Institute of Physics Publishing. All rights reserved.

关键词： Statistical methods

来源：评论

学校读者我要写书评

暂无评论

Platoon-Based Distributed control for Automated Material Handling systems

Platoon-Based Distributed Control for Automated Material Han...

引用

International Conference on control, Decision and Information Technologies (CoDIT)

作者： Alessandro Bozzi Jose-Fernando Jimenez Camilo Hernandez-Rodriguez Eliana-Maria Gonzalez-Neira Damien Trentesaux Department of Informatics Bioengineering Robotics and Systems Engineering (DIBRIS) University of Genova Italy Laboratory System and Materials for Mechatronics (SYMME) Université Savoie Mont Blanc Annecy France Industrial Engineering Department Pontificia Universidad Javeriana Bogota Columbia Laboratory of Industrial and Human Automation Control Mechanical Engineering and Computer Science (LAMIH) Université Polytechnique Hauts-de-France Valenciennes France

Smart manufacturing systems combine collaborative and autonomous control systems to provide optimal performance and respond to volatile demands. A key component of smart manufacturing is material management, as raw materials, work-in-progress, and finished products must be transported to the right location at the right time. However, during the production phase, it is imperative to establish an optimal traffic control scheme in order to avoid congestion and optimize material flow. This paper proposes a platooning-based control procedure implemented in a virtual agent-based manufacturing environment to increase the material flow efficiency on a manufacturing shop floor. An analysis of the production makespan and an indicator of the level of congestion in the material handling system was conducted in order to determine productivity improvements. An agent-based simulation is used to test the effectiveness of the novel platooning approach versus a traditional basic approach. A multivariate analysis of variance (MANOVA) reveals that the inclusion of a platoon-based procedure improves production performance and reduces congestion risks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-Branch Mutual-Distillation Transformer for EEG-Based Seizure Subtype Classification

arXiv

引用

arXiv 2024年

作者： Peng, Ruimin Du, Zhenbang Zhao, Changming Luo, Jingwei Liu, Wenzhong Chen, Xinxing Wu, Dongrui Belt and Road Joint Laboratory on Measurement and Control Technology Huazhong University of Science and Technology Wuhan430074 China AI Platform Software Engineering Research Center Dongfeng Corporation Research & Development Institute Wuhan430072 China China Electronic System Technology Co. Ltd. Beijing100089 China Guangdong Provincial Key Laboratory of Human-Augmentation and Rehabilitation Robotics in Universities Southern University of Science and Technology Shenzhen518055 China

Cross-subject electroencephalogram (EEG) based seizure subtype classification is very important in precise epilepsy diagnostics. Deep learning is a promising solution, due to its ability to automatically extract latent patterns. However, it usually requires a large amount of training data, which may not always be available in clinical practice. This paper proposes Multi-Branch Mutual-Distillation (MBMD) Transformer for cross-subject EEG-based seizure subtype classification, which can be effectively trained from small labeled data. MBMD Transformer replaces all even-numbered encoder blocks of the vanilla Vision Transformer by our designed multi-branch encoder blocks. A mutual-distillation strategy is proposed to transfer knowledge between the raw EEG data and its wavelets of different frequency bands. Experiments on two public EEG datasets demonstrated that our proposed MBMD Transformer outperformed several traditional machine learning and state-of-the-art deep learning approaches. To our knowledge, this is the first work on knowledge distillation for EEG-based seizure subtype classification. © 2024, CC BY.

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Precise LiDAR SLAM in Structured Scene Using Finite Plane and Prior Constraint

Precise LiDAR SLAM in Structured Scene Using Finite Plane an...

引用

IEEE International Conference on robotics and Biomimetics

作者： Yuhui Xie Wentao Zhao Jiahao Wang Jingchuan Wang Weidong Chen Key Laboratory of System Control and Information Processing Ministry of Education Institute of Medical Robotics and Department of Automation Shanghai Jiao Tong University Shanghai China Ministry of Education School of Mechanical Engineering Shanghai Jiao Tong University Shanghai China

ISBN: (纸本)9781665481106

In this paper, we propose a precise LiDAR SLAM in optimization framework using plane-like object as landmark. Compared to general methods, finite plane feature is used to represent landmark and a new residual model is designed, making constraint from edge of landmark can be used to limit the parallel position between LiDAR and landmark, leading to a more accurate result. Moreover, floor plan is used to provide global pose for reduction of drift, and additionally feature orientation prior is used to prevent map distortion when updating the inaccurate part of floor plan. Experiments are conducted using data collected in real environment. The result qualitatively shows that proposed method can build a corrected map from floor plan without distortion, and quantitatively verifies that proposed method can outperform other baseline methods in accuracy.

关键词： Laser radar Simultaneous localization and mapping Image analysis Biomimetics Distortion Floors Robots

来源：评论

学校读者我要写书评

暂无评论

CDFI: Cross Domain Feature Interaction for Robust Bronchi Lumen Detection

CDFI: Cross Domain Feature Interaction for Robust Bronchi Lu...

引用

IEEE International Conference on robotics and Automation (ICRA)

作者： Jiasheng Xu Tianyi Zhang Yangqian Wu Jie Yang Guang–Zhong Yang Yun Gu Institute of Medical Robotics Shanghai Jiao Tong University Shanghai CHINA Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai CHINA MOE Key Laboratory of System Control and Information Processing Shanghai P.R. China Shanghai Center for Brain Science and Brain-Inspired Technology Shanghai China

Endobronchial intervention is increasingly used as a minimally invasive means for the treatment of pulmonary diseases. In order to reduce the difficulty of manipulation in complex airway networks, robust lumen detection is essential for intraoperative guidance. However, these methods are sensitive to visual artifacts which are inevitable during the surgery. In this work, a cross domain feature interaction (CDFI) network is proposed to extract the structural features of lumens, as well as to provide artifact cues to characterize the visual features. To effectively extract the structural and artifact features, the Quadruple Feature Constraints (QFC) module is designed to constrain the intrinsic connections of samples with various imaging-quality. Furthermore, we design a Guided Feature Fusion (GFF) module to supervise the model for adaptive feature fusion based on different types of artifacts. Results show that the features extracted by the proposed method can preserve the structural information of lumen in the presence of large visual variations, bringing much-improved lumen detection accuracy.

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：