检索结果-内蒙古大学图书馆

Chinese Control and Decision Conference, CCDC

作者： Guopeng Zhang Tao Zhang Qingxin Li Li Chen Hualiang Zhang Jingzhe Fang Yiqiang Zhang School of Artificial Intelligence Shenyang University of Technology Shenyang Institute of Automation Chinese Academy of Sciences Shenyang China Key Laboratory of Networked Control Systems Chinese Academy of Sciences Shenyang Institute of Automation Chinese Academy of Sciences Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang China School of Artificial Intelligence Shenyang University of Technology Shenyang China School of Electrical Engineering Shenyang University of Technology Shenyang Institute of Automation Chinese Academy of Sciences Shenyang China

ISBN: (数字)9798350387780

ISBN: (纸本)9798350387797

The X-Y precision planar motion stage is a highly advanced motion control device that has been widely utilized in various precision motion control fields. This platform achieves precise positioning and control of objects through precise motion along the X and Y axes, providing an outstanding solution for scientific research, industrial production, and other applications. The X-Y platform consists of two permanent magnet synchronous linear motors. In the three-loop control of the motor, when the current loop uses a PID controller, there is a problem in tuning the parameters. This paper replaces the original PID controller in the current loop with a model predictive controller. Simultaneously, addressing the issue of improving the tracking accuracy of the X-Y precision planar motion stage, a forgetting high-order variable gain PD-type iterative learning control is proposed. Subsequently, both of these control methods are applied to the Simulink model of the X-Y precision motion stage motor. This not only solves the complex problem of tuning the motor current loop parameters but also improves the tracking accuracy of the X-Y precision motion stage. The results indicate that when using both model predictive controllers and forgetting high-order variable gain PD-type iterative learning controllers for motor control, not only can the tuning of current loop parameters be omitted, but also tracking errors can be effectively reduced, and the system convergence speed can be accelerated.

关键词： Tracking loops Trajectory tracking Predictive models Motors Mathematical models Motion control Tuning

来源：评论

学校读者我要写书评

暂无评论

RealisHuman: A Two-Stage Approach for Refining Malformed Human Parts in Generated Images 39

RealisHuman: A Two-Stage Approach for Refining Malformed Hum...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Wang, Benzhi Zhou, Jingkai Bai, Jingqi Yang, Yang Chen, Weihua Wang, Fan Lei, Zhen State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science & Innovation Chinese Academy of Sciences China Alibaba Group China

ISBN: (纸本)157735897X

In recent years, diffusion models have revolutionized visual generation, outperforming traditional frameworks like Generative Adversarial Networks (GANs). However, generating images of humans with realistic semantic parts, such as hands and faces, remains a significant challenge due to their intricate structural complexity. To address this issue, we propose a novel post-processing solution named RealisHuman. The RealisHuman framework operates in two stages. First, it generates realistic human parts, such as hands or faces, using the original malformed parts as references, ensuring consistent details with the original image. Second, it seamlessly integrates the rectified human parts back into their corresponding positions by repainting the surrounding areas to ensure smooth and realistic blending. The RealisHuman framework significantly enhances the realism of human generation, as demonstrated by notable improvements in both qualitative and quantitative metrics. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis for Polar Codes under Successive Cancellation List Decoding with Fixed List Size

arXiv

引用

arXiv 2023年

作者： Piao, Jinnan Li, Dong Yu, Xueting Li, Zhibo Yang, Ming Liu, Jindi Zeng, Peng The State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Shenyang110016 China The Key Laboratory of Networked Control Systems Chinese Academy of Sciences Shenyang110016 China The Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang110169 China

In this paper, we first indicate that the block error event of polar codes under successive cancellation list (SCL) decoding is composed of path loss (PL) error event and path selection (PS) error event, where the PL error event is that correct codeword is lost during the SCL decoding and the PS error event is that correct codeword is reserved in the decoded list but not selected as the decoded codeword. Then, we simplify the PL error event by assuming the all-zero codeword is transmitted and derive the probability lower bound via the joint probability density of the log-likelihood ratios of information bits. Meanwhile, the union bound calculated by the minimum weight distribution is used to evaluate the probability of the PS error event. With the performance analysis, we design a greedy bit-swapping (BS) algorithm to construct polar codes by gradually swapping information bit and frozen bit to reduce the performance lower bound of SCL decoding. The simulation results show that the BLER performance of SCL decoding is close to the lower bound in the medium to high signal-to-noise ratio region and we can optimize the lower bound to improve the BLER performance of SCL decoding by the BS algorithm. Copyright © 2023, The Authors. All rights reserved.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Learning Object-Centric Representation via Reverse Hierarchy Guidance

arXiv

引用

arXiv 2024年

作者： Zou, Junhong Zhu, Xiangyu Zhang, Zhaoxiang Lei, Zhen State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science & Innovation Chinese Academy of Sciences Hong Kong

Object-Centric Learning (OCL) seeks to enable Neural Networks to identify individual objects in visual scenes, which is crucial for interpretable visual comprehension and reasoning. Most existing OCL models adopt auto-encoding structures and learn to decompose visual scenes through specially designed inductive bias, which causes the model to miss small objects during reconstruction. Reverse hierarchy theory proposes that human vision corrects perception errors through a top-down visual pathway that returns to bottom-level neurons and acquires more detailed information, inspired by which we propose Reverse Hierarchy Guided Network (RHGNet) that introduces a top-down pathway that works in different ways in the training and inference processes. This pathway allows for guiding bottom-level features with top-level object representations during training, as well as encompassing information from bottom-level features into perception during inference. Our model achieves SOTA performance on several commonly used datasets including CLEVR, CLEVRTex and MOVi-C. We demonstrate with experiments that our method promotes the discovery of small objects and also generalizes well on complex real-world scenes. Code will be available at https://***/r/RHGNet-6CEF. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Lidar-Inertial and Downward Vision Based Mobile Robot SLAM in Digital Workshops

Lidar-Inertial and Downward Vision Based Mobile Robot SLAM i...

引用

Data Driven Control and Learning systems (DDCLS)

作者： Yanchao Zhang Hao Geng Bin Liu Yinlong Zhang Xudong Yuan Sichao Zhang School of Information Science and Engineering Shenyang University of Technology Shenyang China Guangzhou Institute of Industrial Intelligence Guangzhou China State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Shenyang China Key Laboratory of Networked Control Systems Chinese Academy of Sciences Shenyang China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang China

ISBN: (数字)9798350361674

ISBN: (纸本)9798350361681

In order to solve the problems of long-term cumulative error and the absence of absolute pose in the existing Lidar-Inertial Measurement Unit (IMU) fusion methods for Automated Guided Vehicle (AGV) Simultaneous Localization and Mapping (SLAM), we propose a new SLAM method for mobile robot, integrating Lidar, IMU, and a downward vision camera. Firstly, we tightly couple Lidar-IMU and downward vision camera using a multimodal fusion framework. Downward visual measurements are integrated as QR factors into the factor graph, providing absolute pose information to correct cumulative error in the Lidar-IMU system and enable the joint pose estimation of the AGV. Secondly, an enhanced Iterative Closest Point (ICP) point cloud matching algorithm has been developed to correct the incorrect pose. This approach achieves Scan-to-Map alignment by employing a coarse-to-fine matching strategy, thereby enhancing the accuracy of point cloud alignment. Finally, the extended Kalman filter is used to optimize the Lidar pose drift, as well as the Lidar mapping effectively by using the Lidar-IMU as the measurements and the downward visual measurements as the updates. The algorithm has been extensively evaluated on the developed AGV experimental platform. The experimental results show that the position error is less than 0.02 meters and the orientation error is less than 1°, which solves the problems of long-term error accumulation and the lack of absolute pose information for AGV navigation.

关键词： Point cloud compression Visualization Laser radar Simultaneous localization and mapping Conferences Robot vision systems Measurement uncertainty

来源：评论

学校读者我要写书评

暂无评论

Optimizing Handwritten Character Recognition systems: Neural Networks vs. Statistical Methods

Optimizing Handwritten Character Recognition Systems: Neural...

引用

Intelligent systems and Advanced Applications (ICISAA), International Conference on

作者： Vikash K. Agrawal Srinivasa Rao Bogireddy Lalit N. Patil Mahesh Sonekar Yashraj M. Patil Vikas Singh Panwar Department of Automation and Robotics Engineering Dr. D. Y. Patil Institute of Technology Pune India Horizon Systems Inc Phoenix AZ USA Department of Mechanical Engineering Dr. D. Y. Patil Institute of Technology Pune India

ISBN: (数字)9798331539948

ISBN: (纸本)9798331539955

This investigation focuses on the identification of handwritten characters using neural network technology combined with statistical methods. Handwriting recognition has gained popularity due to its wide range of applications in document processing, signature authentication, and digitized communication. The use of neural networks, particularly deep learning models, has greatly enhanced the accuracy and efficiency of character recognition systems. This study provides a comprehensive analysis aimed at evaluating the effectiveness of neural networks in recognizing handwritten characters. Various neural network architectures, including convolution and recurrent neural networks, are explored; each designed to address specific challenges in character recognition. Additionally, statistical methods are employed for feature extraction, further improving recognition accuracy. Extensive testing and benchmarking are conducted on diverse handwritten datasets to assess the performance of the proposed approaches. The results demonstrate the significant benefits of combining artificial neural networks with statistical techniques to achieve high levels of accuracy in character recognition, paving the way for practical applications. This study makes substantial contributions to the field of handwritten character recognition by highlighting the synergistic advantages of neural networks and statistical methods, which will aid in the development of more accurate and robust authentication systems for future applications.

关键词： Training Handwriting recognition Accuracy Statistical analysis Neural networks Authentication Artificial neural networks Writing Feature extraction Character recognition

来源：评论

学校读者我要写书评

暂无评论

System Design and Workspace Optimization of a Parallel Mechanism-Based Portable Robot for Remote Ultrasound

System Design and Workspace Optimization of a Parallel Mecha...

引用

automation, Control and robotics Engineering (CACRE), International Conference on

作者： Zhaokun Deng Xilong Hou Mingrui Hao Shuangyi Wang The State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences Beijing China Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science & Innovation Chinese Academy of Sciences Hong Kong China

The robotic ultrasound system has the potential to improve the conventional practice of diagnosing. Because of the adequate degrees of freedom embedded in a small footprint, the parallel mechanism-based ultrasound robot has attracted attention in the field. However, the analysis of its configuration, design parameters, and workspace is limited. To solve this issue and further promote the potential clinical translation, this paper proposes a task-driven, two-stage mechanism optimization method using the effective regular workspace and the local condition index to determine the parameters for the demanding clinic workspace of a parallel mechanism-based ultrasound robot. The design and implementation method of the robot are then introduced, along with the justification of parameter selection. To analyze the performance, an optical tracking-based experiment and a phantom-based human-robot comparison study were performed. The results show that the workspace meets the required clinical needs, and despite its small footprint, the mechanism could have a reasonable workspace. The kinematic error was found to be 0.2 mm and 0.3°. Based on the above results and the quantitative analysis of the ultrasound images acquired manually and robotically, it was concluded that the robot can effectively deliver the demand function and would be a promising tool for further deployment.

关键词：

来源：评论

学校读者我要写书评

暂无评论

MvEchoSeg: One-shot In-context Learning for Multi-view Echocardiography Segmentation

MvEchoSeg: One-shot In-context Learning for Multi-view Echoc...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Guo, Yuting Duan, Ya Song, Wenfeng Li, Nannan Li, Aili Li, Shuai Beihang University State Key Laboratory of Virtual Reality Technology and Systems Beijing100191 China University of Chinese Academy of Sciences State Key Laboratory of Robotics Shenyang Institute of Automation Chinese Academy of Sciences Shenyang110016 China Beijing Information Science and Technology University School of Computer Science Beijing100192 China Dalian Maritime University School of Information Science and Technology Dalian116000 China China-Japan Friendship Hospital Beijing100029 China

ISBN: (纸本)9798350386226

Echocardiography is the clinical standard for evaluation of cardiac morphology, and function, and providing hemodynamic parameters in patients with known or suspected heart disease. Due to the diversity and complexity of diagnostic tasks, comprehensive interpretation of echocardiography often requires multi-view imaging for integrated metric analysis. Deep learning methods have become the mainstream approach for echocardiography segmentation. However, achieving segmentation of multi-view echocardiography still requires a substantial amount of annotation. To remedy this, we propose a one-shot in-context learning network mvEchoSeg for multi-view echocardiography segmentation. This network requires only one annotated image for each view. Specifically, we propose a Task Prompt Identifier (TPI) module to identify the task type of the image and allocate the most precise task prompt for it, with minimal adaptation of CLIP and few-shot strategies. Additionally, we leverage a unified In-Context Model (ICM) capable of performing a diverse set of echocardiography segmentation tasks automatically. Using fine-tuning and low-rank adapters improved the performance of the pre-trained model, achieving significant results with minimal training cost. Furthermore, we collect a multi-view echocardiography dataset (MVECD) with 8 views to evaluate our method. The results show an improvement of more than 10% in the DICE score compared to the SOTA foundational medical image segmentation models. To our knowledge, this is the first exploration of a one-shot model for multi-view echocardiography segmentation. Our codes and models are available at https://***/stellating/mvEchoSeg. © 2024 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

WindowMixer: Intra-Window and Inter-Window Modeling for Time Series Forecasting

arXiv

引用

arXiv 2024年

作者： Liu, Quangao Li, Ruiqi Jiang, Maowei Yang, Wei Liang, Chen Pang, LongLong Zou, Zhuozhang Key Laboratory of Networked Control Systems Chinese Academy of Sciences Shengyang110016 China Shenyang Institute of Automation Chinese Academy of Sciences China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences China University of Chinese Academy of Sciences China

Time series forecasting (TSF) is widely applied in various fields such as economic forecasting, weather prediction, traffic flow analysis, and public health surveillance. The core of TSF is to capture the underlying temporal patterns in the data. However, real-world time series data exhibit complex temporal patterns and may include additional noise, outliers, missing values, etc., making accurate forecasting highly challenging. Current mainstream methods typically naively treat each time point as an isolated entity, attempting to model point-to-point relationships and uncover temporal patterns within the data, which has limitations when dealing with complex time series data. Specifically, point-by-point modeling fails to take advantage of the abundant contextual information in time series data. This results in a reduced ability to capture temporal patterns and an increased susceptibility to noise. To address these issues, we examine temporal variations from a window-based perspective and present the WindowMixer model, which is built on an all-MLP framework. WindowMixer stems from a simple observation: Time series are recorded in a continuous manner, and the information at any given time point relies on the preceding and succeeding time points for a complete representation. WindowMixer decomposes time series into trend and seasonal components via a sequence decomposition and fusion architecture, handling them individually. For the trend component, we employ a straightforward fully connected (FC) layer to make predictions. For the seasonal component, we first introduce a time window at each time point and project each time window to produce window tokens. Then, we apply the Intra-Window-Mixer and Inter-Window-Mixer to further process these tokens. The Intra-Window-Mixer models relationships within each window token, while the Inter-Window-Mixer models relationships between window tokens. These dual modules allows us to capture more intricate patterns and long-range de

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

A Multi-Model Fusion Approach for Transformer Fault Detection 4

A Multi-Model Fusion Approach for Transformer Fault Detectio...

引用

4th IEEE International Conference on Information Technology, Big Data and Artificial Intelligence, ICIBA 2024

作者： Zhou, Ke Zhang, Wei Li, Ligang Pan, Shaoming Huang, Chengwei Electric Power Research Institute of Guangxi Power Grid CO. LTD. Guangxi Power Grid Equipment Monitoring and Diagnosis Technology Innovation Center Guangxi Key Laboratory of Intelligent Control and Maintenance of Power Equipment Nanning530023 China Shenyang Institute of Automation Chinese Academy of Sciences State Key Laboratory of Robotics Shenyang110016 China Chinese Academy of Sciences Key Laboratory of Networked Control Systems China Power Supply Bureau of Guangxi Hezhou Power Grid Co. Ltd. China

ISBN: (纸本)9798350363586

Addressing the weak generalization capability and suboptimal prediction performance of single mechanistic and data-driven models, this paper proposes a transformer fault detection method based on the fusion of mechanism and data models. Firstly, vibration and current data are collected on-site to train the SVM vibration model and the Adaboost current model. Secondly, current simulation data is obtained using laboratory simulation software. This simulation data, combined with the current data collected under the same conditions on-site, is used as input to train the GBDT error compensation model. Finally, these three models are used as inputs to the GRNN, completing the overall model design. This fusion of mechanistic and data-driven models not only addresses the parameter mismatch issue of the mechanistic model under complex working conditions but also deeply mines time-series data. Experimental results show that The mean squared error of the fusion model is 0.002 lower than that of the data model, and the coefficient of determination is improved by 5.07% compared to the mechanistic model, enhancing the overall predictive accuracy of the model. © 2024 IEEE.

关键词： Data accuracy

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：