检索结果-内蒙古大学图书馆

SSRN 2023年

作者： Wang, Quan Wang, Fei Li, Zhongheng Wang, Zheng Nie, Feiping Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Shaanxi Xi’an710049 China Beijing Key Laboratory of Intelligent Space Robotic Systems Technology and Applications Beijing Institute of Spacecraft System Engineering Beijing100094 China School of Software Engineering Xi’an Jiaotong University Xi’an710049 China Northwestern Polytechnical University Shaanxi Xi’an710072 China

Joint clustering and dimensionality reduction methods are a promising solution to clustering due to its scalability to high-dimensional data. Some methods leverage trace ratio criterion and attain clusters by borrowing the K-means algorithm. However, trace ratio criterion has no close-formed solution for the discriminative projection matrix and the K-means algorithm has a limited capacity to handle the many-cluster problem. In this paper, Coordinate Descent Clustering with Trace Difference criterion (CDCTD) is proposed for joint clustering and feature extraction. Formulating the objective function as a direct trace difference criterion containing a balance parameter, CDCTD harmonizes between-cluster scatter maximization and within-cluster scatter minimization by the balance parameter. Using the direct trace difference criterion, CDCTD can straightforward solve for the discriminative projection matrix and avoid obtaining a poor discriminative projection matrix in the iterative manner when a bad cluster start is given. CDCTD uses the coordinate descent method for clustering, improving the ability to address the many-cluster problem. Extensive experiments show that CDCTD has achieved significant performance improvements compared to previous trace ratio criterion related joint clustering and feature extraction methods, and also outperformed other clustering methods in most cases. © 2023, The Authors. All rights reserved.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Rehabilitation Training of Spinal Cord Injured Individuals by a Self-Balancing Exoskeleton Robot

Rehabilitation Training of Spinal Cord Injured Individuals b...

引用

IEEE International Conference on Real-time Computing and robotics (RCAR)

作者： Wentao Li Jinke Li Yong He Dingkui Tian Feng Li Jianquan Sun Xinyu Wu Guangdong Provincial Key Lab of Robotics and Intelligent System Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China SIAT CUHK Joint Laboratory of Robotics and Intelligent Systems Shenzhen College of Advanced Technology University of Chinese Academy of Sciences Shenzhen China China Construction Third Bureau First Engineering MEP Co. Ltd. Shenzhen China

Physical therapy plays an important role in the rehabilitation for spinal cord injured individuals, and exoskeleton can assist with this process. In this paper, a self-balancing exoskeleton robot is introduced, which can provide assist not only in gait training, but also in balance training. In addition, it allows individuals walk with it without other assistive devices. This exoskeleton is designed to have a high degree of compatibility with humans. The structure and control system of the exoskeleton is introduced and the forward and inverse kinematic are solved, which enables the robot to perform various rehabilitation training actions as required. Experiments with a male subject have shown that this exoskeleton achieves the predicted results in gait training and balance training.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Real-time Framework for UAV Indoor Self-Positioning and 3D Mapping Base on 2D Lidar, Stereo Camera and IMU

A Real-time Framework for UAV Indoor Self-Positioning and 3D...

引用

IEEE International Conference on Real-time Computing and robotics (RCAR)

作者： Lebin Liang Haotian Rao Guohao Shen Can Wang Xinyu Wu Guangdong Provincial Key Lab of Robotics and Intelligent System Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China University of Chinese Academy of Sciences Beijing China College of Electronic Engineering (College of Artificial Intelligence) South China Agricultural University Guangzhou China

This paper proposed a real-time Unmanned Aerial Vehicle (UAV) autonomous localization framework based on onboard sensors, including 2D Lidar, Stereo Camera and Inertial Measurement Unit (IMU), built on the open-source VINS-Fusion software to achieve real-time and drift-free UAV pose estimation. The proposed framework eliminates the motion distortion of Lidar during UAV motion by combining Visual-Inertial Odometry (VIO). It obtains global localization information of Lidar by PL-ICP algorithm based on a global point cloud map. Graph optimization method is used to fuse the local positional estimation and the global information from the Lidar, which eliminate the cumulative error caused by VIO. The feasibility of the method is verified by experiments and compared with the method of only running Visual-Inertial Navigation systems (VINS), the framework improves the accuracy of localization. In addition, we have implemented the construction of octree maps of 3D environments through the framework of this system.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Fast automatic docking control for electric tractors

引用

Fundamental Research 2024年

作者： Li, Pengpeng Bai, Yu Li, Weihua Chen, Bowen You, Bindi Zhuang, Hongchao Wang, Jianfeng Gao, Haibo Deng, Zongquan School of Astronautics Harbin Institute of Technology Harbin 150001 China School of Automotive Engineering Harbin Institute of Technology (Weihai) Weihai 264201 China Yangtze River Delta HIT robot technology research institute Wuhu 241060 China State Key Laboratory of Robotics and System Harbin Institute of Technology Harbin 150001 China Tianjin University of Technology and Education Tianjin 300222 China

In this study, we investigate the automatic docking of an electric tractor with a semi-trailer in narrow and long-distance scenarios. We propose an automatic docking method based on a multi-sensor fusion high-precision positioning algorithm and a multi-strategy docking control algorithm. To solve the heading-angle error of the involved camera due to mechanical errors, a compensation value in the form of angle deviation is introduced. To alleviate the increase in the camera detection error with longitudinal distance, a piecewise coordinate compensation function is proposed. Repositioning is performed using light detection and ranging (Lidar) to improve the pose-perception accuracy to the millimeter level. To avoid the interference of point clouds on the relocation accuracy of Lidar, a search-and-location method that combines both small and large neighborhoods is proposed. To achieve the desired docking success rate, a multi-strategy-adjustment control algorithm is used to enable comprehensive adjustment, including approximate control of the maximum angle based on lateral deviation and fine proportional-integral-differential (PID) control based on lateral deviation and longitudinal distance, security check cycle and premature warning mechanism, and risk avoidance control. Finally, the reliability and stability of the design scheme are verified by conducting multiple experiments involving actual vehicles. © 2024

关键词： Automatic docking Electric tractor Fusion positioning Piecewise proportional-integral-differential

来源：评论

学校读者我要写书评

暂无评论

Detect and attribute the extreme maize yield losses based on spatio-temporal deep learning

引用

Fundamental Research 2023年第6期3卷 951-959页

作者： Renhai Zhong Yue Zhu Xuhui Wang Haifeng Li Bin Wang Fengqi You Luis F.Rodríguez Jingfeng Huang K.C.Ting Yibin Ying Tao Lin College of Biosystems Engineering and Food Science Zhejiang UniversityHangzhouZhejiang 310058China International Campus Zhejiang UniversityHainingZhejiang 314400China Sino-French Institute for Earth System Science College of Urban and Environmental SciencesPeking UniversityBeijing 100871China School of Geosciences and Info-Physics Central South UniversitySouth Lushan RoadChangsha 410000China NSW Department of Primary Industries Wagga Wagga Agricultural InstitutePine Gully Road Wagga WaggaNSW 2650Australia Robert Frederick Smith School of Chemical and Biomolecular Engineering Cornell UniversityIthacaNY 14853USA Department of Agricultural and Biological Engineering University of Illinois at Urbana-ChampaignUrbanaIL 61801USA Institute of Applied Remote Sensing and Information Technology Zhejiang UniversityHangzhouZhejiang 310058China Key Laboratory of On Site Processing Equipment for Agricultural Products Ministry of Agriculture and Rural AffairsHangzhouZhejiang 310058China Key Laboratory of Intelligent Equipment and Robotics for Agriculture of Zhejiang Province HangzhouZhejiang 310058China

Providing accurate crop yield estimations at large spatial scales and understanding yield losses under extreme climate stress is an urgent challenge for sustaining global food *** the data-driven deep learning approach has shown great capacity in predicting yield patterns,its capacity to detect and attribute the impacts of climatic extremes on yields remains *** this study,we developed a deep neural network based multi-task learning framework to estimate variations of maize yield at the county level over the US Corn Belt from 2006 to 2018,with a special focus on the extreme yield loss in *** found that our deep learning model hindcasted the yield variations with good accuracy for 2006-2018(R^(2)=0.81)and well reproduced the extreme yield anomalies in 2012(R^(2)=0.79).Further attribution analysis indicated that extreme heat stress was the major cause for yield loss,contributing to 72.5%of the yield loss,followed by anomalies of vapor pressure deficit(17.6%)and precipitation(10.8%).Our deep learning model was also able to estimate the accumulated impact of climatic factors on maize yield and identify that the silking phase was the most critical stage shaping the yield response to extreme climate stress in *** results provide a new framework of spatio-temporal deep learning to assess and attribute the crop yield response to climate variations in the data rich era.

关键词： Crop yield estimation Deep Learning Long short-term memory Multi-task learning Extreme yield loss Attribution analysis

来源：评论

学校读者我要写书评

暂无评论

Survey on Large Language Model-Enhanced Reinforcement Learning: Concept, Taxonomy, and Methods

引用

IEEE Transactions on Neural Networks and Learning systems 2024年第6期36卷 9737-9757页

作者： Yuji Cao Huan Zhao Yuheng Cheng Ting Shu Yue Chen Guolong Liu Gaoqi Liang Junhua Zhao Jinyue Yan Yun Li Department of Mechanical and Automation Engineering The Chinese University of Hong Kong Hong Kong SAR China Department of Building Environment and Energy Engineering The Hong Kong Polytechnic University Hong Kong China School of Science and Engineering The Chinese University of Hong Kong Shenzhen China Center for Crowd Intelligence Shenzhen Institute of Artificial Intelligence and Robotics for Society (AIRS) Shenzhen China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen China School of Electrical and Electronic Engineering Nanyang Technological University Jurong West Singapore School of Mechanical Engineering and Automation Harbin Institute of Technology Shenzhen China Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China Shenzhen China i4AI Ltd. London U.K.

With extensive pretrained knowledge and high-level general capabilities, large language models (LLMs) emerge as a promising avenue to augment reinforcement learning (RL) in aspects, such as multitask learning, sample efficiency, and high-level task planning. In this survey, we provide a comprehensive review of the existing literature in LLM-enhanced RL and summarize its characteristics compared with conventional RL methods, aiming to clarify the research scope and directions for future studies. Utilizing the classical agent-environment interaction paradigm, we propose a structured taxonomy to systematically categorize LLMs’ functionalities in RL, including four roles: information processor, reward designer, decision-maker, and generator. For each role, we summarize the methodologies, analyze the specific RL challenges that are mitigated and provide insights into future directions. Finally, the comparative analysis of each role, potential applications, prospective opportunities, and challenges of the LLM-enhanced RL are discussed. By proposing this taxonomy, we aim to provide a framework for researchers to effectively leverage LLMs in the RL field, potentially accelerating RL applications in complex applications, such as robotics, autonomous driving, and energy systems.

关键词： Taxonomy Surveys Visualization Reviews Planning Reinforcement learning Games Transformers Natural language processing Large language models

来源：评论

学校读者我要写书评

暂无评论

Multi-scale Contrastive Learning for Gastroenteroscopy Classification

Multi-scale Contrastive Learning for Gastroenteroscopy Class...

引用

Annual IEEE Symposium on Computer-Based Medical systems

作者： Dan Li Xuechen Li Zhibin Peng Wenting Chen Linlin Shen Guangyao Wu Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University National Engineering Laboratory for Big Data System Computing Technology ShenZhen University Shenzhen China City University of Hong Kong Hong Kong SAR China Shenzhen Institute of Artificial Intelligence & Robotics for Society Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University General Hospital

In gastroenteroscopy image analysis, numerous CADs demonstrate that deep learning aids doctors' diagnosis. The shapes and sizes of the lesions are varied. And in the clinic, the dataset appears to be data imbalanced. However, existing methods directly classify by texture and ignore lesions with various shapes and sizes. To address the issue above, we propose a deep neural network, which consists of multi-scale feature extraction, contrastive feature learning and a multi-scale feature fusion module. We train the contrastive feature learning module and multi-scale feature fusion module simultaneously to alleviate the issue of data distribution differences. Thus, the proposed network can better identify various categories. Extensive experiments on the Hyper Kvasir dataset show that the proposed Hybrid-M2CL outperforms the benchmark proposed by the dataset with 5.0% Macro Precision, 3.3% Macro Recall, 3.4% Macro F1-score, 3.3% Micro Precision, 3.6% MCC. In addition, it outperforms the SOTA by 1.1% Macro F1-score, 2.6% MCC, and 2.0% B-ACC.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Fractional Kelvin-Voigt based Viscoelastic Control for self-Balancing Exoskeleton

Fractional Kelvin-Voigt based Viscoelastic Control for self-...

引用

IEEE International Conference on robotics and Biomimetics

作者： Wentao Li Feng Li Ming Yang Jinke Li Ming Li Yuanpei Zhu Meng Yin Dingkui Tian Xinyu Wu Department of Electronic and Electrical Engineering Southern University of Science and Technology Shenzhen China Chinese Academy of Sciences Shenzhen Institute of Advanced Technology Shenzhen China Guangdong Provincial Key Lab of Robotics and Intelligent System Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen China University of Science and Technology of China Hefei China

Self-balancing exoskeleton robots enable hands-free walking for the wearer, and the problem of its stable walking has been a challenge. In this paper, to achieve compliant and stable walking of a self-balancing exoskeleton robot and reduce the shock between the robot and the ground, we propose a new fractional kelvin voigt viscoelastic controller (FKVC). The proposed FKVC incorporates the advantages of fractional order and the kelvin voigt viscoelastic to endow the self-balancing exoskeleton with stable and dynamic walking capabilities. Simulation results show that the actual CoM position of the exoskeleton well tracks the reference CoM position, and the actual ZMP converges to the reference ZMP position. The ZMP is always within the effective support region of the biped. Therefore, we experimentally proved the practicality and effectiveness of the FKVC method, and the self-balancing exoskeleton successfully realizes stable dynamic walking.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Time-Optimal Trajectory Planning Based on Dynamics for Industrial Robot 5

Time-Optimal Trajectory Planning Based on Dynamics for Indus...

引用

5th International Conference on Automation, Control and robotics engineering, CACRE 2020

作者： Wang, Rundong Kong, Minxiu Cheng, Yanchun Castellani, Marco Harbin Institute of Technology State Key Laboratory of Robotics and System School of Mechatronics Engineering Harbin China University of Birmingham Department of Mechanical Engineering Birmingham United Kingdom

ISBN: (纸本)9781728198880

Given the trajectory of the industrial robot in Cartesian space, the existing methods only consider the kinematic constraints when optimizing the time trajectory planning, and ignore the maximum driving ability of the motor, which is likely to cause an overrun alarm and greatly reduce the trajectory accuracy. Therefore, this paper studies the time-optimal trajectory planning algorithm based on dynamics. Firstly, the algorithm discretizes the path, calculates the Cartesian coordinates of each discrete point and the differential about the path length. Then, this algorithm considers the constraints of dynamics and kinematics, uses the principle of linear programming, recursively calculates the upper limit value of speed and the maximum acceleration value at each discrete point, and calculates the speed value at each discrete point and keeps it within its allowable ranges. Finally, this algorithm uses the speed to calculate the corresponding time of each discrete point, and according to the interpolation period, outputs the path position of each period, the position and orientation in the Cartesian space and joint parameters. The MATLAB program is written to verify the feasibility and efficiency of the algorithm. © 2020 IEEE.

关键词： Industrial robots

来源：评论

学校读者我要写书评

暂无评论

Fragment Distance-Guided Dual-Stream Learning for Automatic Pelvic Fracture Segmentation

SSRN

引用

SSRN 2024年

作者： Zeng, Bolun Wang, Huixiang Joskowicz, Leo Chen, Xiaojun Institute of Biomedical Manufacturing and Life Quality Engineering State Key Laboratory of Mechanical System and Vibration School of Mechanical Engineering Shanghai Jiao Tong University Shanghai China Department of Orthopedics National Center for Orthopedics Shanghai Sixth People’s Hospital Shanghai Jiao Tong University School of Medicine Shanghai China School of Computer Science and Engineering Edmond and Lily Safra Center for Brain Sciences Hebrew University of Jerusalem Jerusalem Israel Institute of Medical Robotics Shanghai Jiao Tong University Shanghai China

Pelvic fracture is a complex and severe injury. Accurate diagnosis and treatment planning require the segmentation of the pelvic structure and the fractured fragments from preoperative CT scans. However, this segmentation is a challenging task, as the fragments from a pelvic fracture typically exhibit considerable variability and irregularity in the morphologies, locations, and quantities. In this study, a novel dual-stream learning framework is proposed for the automatic segmentation and category labeling of pelvic fractures. First, we develop a dual-branch architecture leveraging distance learning from bone fragments for enhanced spatial information, aiding in fracture segmentation and improving the identification of bone fragments with variable positioning. Second, we implement a multi-size feature fusion module that adaptively aggregates features from diverse receptive fields tailored to targets of different sizes and shapes, thus boosting segmentation performance. Extensive experiments on two pelvic fracture datasets from different medical centers demonstrated the accuracy and generalizability of the proposed method. It achieves a mean Dice coefficient and mean F1-score of 0.935[[EQUATION]]0.068 and 0.930[[EQUATION]]0.057 in the dataset FracCLINIC, and 0.955[[EQUATION]]0.072 and 0.912[[EQUATION]]0.129 in the dataset FracSegData, which are superior than other comparing methods. Our method optimizes the process of pelvic fracture segmentation, potentially serving as an effective tool for preoperative planning in the clinical management of pelvic fractures. © 2024, The Authors. All rights reserved.

关键词： Computerized tomography

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：