检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Nona Rajabi Antônio H. Ribeiro Miguel Vasco Danica Kragic Intelligent Systems KTH Royal Institute of Technology Stockholm Sweden Information Technology Uppsala University Uppsala Sweden

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Cross-validation is commonly used to estimate machine learning model performance on new samples. However, using it for both hyperparameter selection and error estimation can lead to overestimating model performance, especially with extensive hyperparameter searches that overly tailor models to validation data. We demonstrate that deep learning further amplifies this bias, with even minor model adjustments causing significant overestimation. Our extensive experiments on simulated and real data focus on the bias from early stopping during cross-validation. We find that overestimation intensifies with network depth and is especially severe in small datasets, which are common in physiological signal processing applications. Selecting the early stopping point during cross-validation can result in ROC-AUC estimates exceeding 90% on random data, and this effect persists across various sample sizes, architectures, and network sizes 1 .

关键词： Deep learning Error analysis Signal processing Data models Physiology Acoustics Speech processing

来源：评论

学校读者我要写书评

暂无评论

A Balance Control Method for Wheeled Bipedal Robot Based on Reinforcement Learning

A Balance Control Method for Wheeled Bipedal Robot Based on ...

引用

第42届中国控制会议

作者： Jing Zhang Biao Lu Zhongyan Liu Yonghui Xie Wankai Chen Weijie Jiang Institute of Robotics and Automatic Information System Nankai University Institute of Intelligence Technology and Robotics Systems Shenzhen Research Institute of Nankai University

Wheeled bipedal robot(WBR) fully combines the advantages of wheeled robot and legged robot, which enables it to adapt to more complex terrains and complete more challenging tasks. Therefore, WBR has been widely researched in recent years. However, at the same time, due to its under-actuation, nonlinearity, and instability characteristics, the balance control of WBR is a difficult problem. In this paper, a reinforcement learning(RL) algorithm based on path integral is ***, the relationship between path integral and stochastic optimal control is analyzed, and the numerical solution of stochastic dynamical systems is obtained. Second, the solution framework is extended to a parameter update strategy of RL. Since the matrix inversion and gradient calculation are avoided during the strategy update process, the training is accelerated. Third,a model-based Lyapunov method is introduced to parameterize the control variables, which not only ensures the stability of the system in training, but also guarantees good performance of the system when applying simulation results to actual *** results show that the proposed method has satisfactory balance control performance and can effectively resist external disturbances.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dynamics Modeling and Controller Design for a Novel Indoor Blimp

Dynamics Modeling and Controller Design for a Novel Indoor B...

引用

第42届中国控制会议

作者： Jinyang Dong Yongchun Fang Institute of Robotics and Automatic Information System Nankai University Institute of Intelligence Technology and Robotics Systems Shenzhen Research Institute of Nankai University

Due to its safety and endurance, the indoor blimp robot is an ideal choice for cruise monitoring and inspection. In order to adapt to the complex indoor scene, a helium-filled blimp is studied in this paper. The dynamic model of the aerial robot is established by using Newton Euler equation, and the main characteristics are then analyzed based on the derived ***, the linear quadratic regulator controller is designed to control the airship motion in the vertical plane. Based on the model, a simulation platform is constructed for further analysis and controller verification. Finally, simulation results are presented to show the correctness of the dynamic model.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Facial expression recognition under occlusion conditions based on multi-feature cross-attention

引用

Journal of intelligent and Fuzzy systems 2024年第4期46卷 9841-9856页

作者： Guo, Sheng Tan, Mian Cai, Shan Zhang, Zaijun Liang, Yihui Feng, Hongxi Zou, Xue Wang, Lin School of Data Science and Information Engineering Guizhou Key Laboratory of Pattern Recognition and Intelligent System Guizhou Minzu University Guiyang China School of Mathematics and Statistics Qiannan Normal University for Nationalities Key Laboratory of Complex Systems and Intelligent Optimization of Guizhou Province Duyun China Zhongshan Institute University of Electronic Science and Technology of China Zhongshan China Guizhou Aerospace Tianma Electromechanical Technology Co LTD Zunyi China

Although facial expression recognition (FER) has a wide range of applications, it may be difficult to achieve under local occlusion conditions which may result in the loss of valuable expression features. This issue has motivated the present study, as a part of which an effective multi-feature cross-attention network (MFCA-Net) is proposed. The MFCA-Net consists of a two-branch network comprising a multi-feature convolution module and a local cross-attention module. Thus, it enables decomposition of facial features into multiple sub-features by the multi-feature convolution module to reduce the impact of local occlusion on facial expression feature extraction. In the next step, the local cross-attention module distinguishes between occluded and unoccluded sub-features and focuses on the latter to facilitate FER. When the MFCA-Net performance is evaluated by applying it to three public large-scale datasets (RAF-DB, FERPlus, and AffectNet), the experimental results confirm its good robustness. Further validation is performed on a real FER dataset with local occlusion of the face. © 2024 – IOS Press.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Semi-Autonomous Navigation Based on Local Semantic Map for Mobile Robot

引用

Journal of Shanghai Jiaotong university(Science) 2025年第1期30卷 27-33页

作者： ZHAO Yanfei XIAO Peng WANG Jingchuan GUO Rui Department of Automation Institute of Medical RoboticsShanghai Jiao Tong UniversityShanghai200240China State Key Laboratory of System Control and Information Processing Ministry of Education of ChinaShanghai200240China Shanghai Engineering Research Center of Intelligent Control and Management Shanghai200240China State Grid Intelligence Technology Co. Ltd.Jinan250013China

Mobile robots represented by smart wheelchairs can assist elderly people with mobility *** paper proposes a multi-mode semi-autonomous navigation system based on a local semantic map for mobile robots,which can assist users to implement accurate navigation(e.g.,docking)in the environment without prior *** order to overcome the problem of repeated oscillations during the docking of traditional local path planning algorithms,this paper adopts a mode-switching method and uses feedback control to perform docking when approaching semantic *** last,comparative experiments were carried out in the real *** show that our method is superior in terms of safety,comfort and docking accuracy.

关键词： semi-autonomous navigation mobile robot semantic map voice interaction

来源：评论

学校读者我要写书评

暂无评论

Minimizing Data Collection Latency for Coexisting Time-Critical Wireless Networks with Tree Topologies

引用

IEEE Transactions on Network and Service Management 2025年第3期22卷 2354-2369页

作者： Zhang, Jialin Liang, Wei Yang, Bo Shi, Huaguang Liang, Ying-Chang Chinese Academy of Sciences State Key Laboratory of Robotics Shenyang Institute of Automation Shenyang110016 China Chinese Academy of Sciences Key Laboratory of Networked Control System Shenyang Institute of Automation Shenyang110016 China Northwest A and F University College of Information Engineering Yangling712100 China Henan University School of Artificial Intelligence Zhengzhou450046 China University of Electronic Science and Technology of China Center for Intelligent Networking and Communications Chengdu611731 China

Time-Critical Wireless Network (TCWN) is a promising communication technology that can satisfy the low latency, high reliability, and deterministic requirements of mission-critical applications. Multiple TCWNs required by various applications inevitably coexist with each other. Most existing works aim to achieve acceptable latency or consider the simplest topology (i.e., line topology). As latency requirements become more stringent, exploring the minimum data collection latency becomes an interesting problem. In this paper, the coexisting system consists of multiple tree-topology-based TCWNs. We first establish a conversion framework to convert an arbitrary tree topology into multiple analogous line topologies to reduce the analysis complexity. We then propose a Time-Critical wireless network Scheduling (TCS) algorithm to minimize the data collection latency of coexisting TCWNs. The TCS algorithm consists of two phases. In the internetwork scheduling phase, we strictly derive a general expression to characterize the practical network requirements. In the intranetwork scheduling phase, we design two levels of priority assignment algorithms to accurately characterize the critical states and resource requirements of different nodes. We conduct extensive simulations to verify the effectiveness of the TCS algorithm. The evaluation results show that the TCS algorithm can achieve minimum data collection latency in more than 99.956% cases, and the maximum difference compared to the optimal value is one time slot. © 2004-2012 IEEE.

关键词： Scheduling algorithms

来源：评论

学校读者我要写书评

暂无评论

Non-contact measurement algorithm of load swing angle for bridge crane based on contour detection

Non-contact measurement algorithm of load swing angle for br...

引用

第42届中国控制会议

作者： Zhongyan Liu Biao Lu Jing Zhang Institute of Robotics and Automatic Information System Nankai University Institute of Intelligence Technology and Robotics Systems Shenzhen Research Institute of Nankai University

The traditional measurement method of the load swing angle for the bridge crane is easily disturbed by the environmental noise, which leads to inaccurate measurement and delayed feedback. In this paper, the computer vision method is used to realize the non-contact measurement of the load swing angle. A rectangular light source is used as a marker and it is detected using an industrial camera. In order to improve the accuracy and real-time performance of the detection, the method of color separation combined with rectangular contour detection is used to quickly identify the marker. Then the swing angle is calculated according to the recognition results by using the camera geometry. It has been demonstrated by experiments that, the algorithm can quickly identify the marker and can accurately calculate the swing angle. In particular, the proposed method has strong independence and anti-interference ability, which is very suitable for bridge crane system.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A MATLAB Multi-body Simulation Platform Toward Multiple Flapping Wing Vehicles

A MATLAB Multi-body Simulation Platform Toward Multiple Flap...

引用

第42届中国控制会议

作者： Jifu Yan Fan Jia Chen Qian Yongchun Fang Institute of Robotics and Automatic Information System Nankai University Institute of Intelligence Technology and Robotics Systems Shenzhen Research Institute of Nankai University

The article notices rare simulation implementations of multiple flapping wing vehicles(FWVs) with the complicate mechanisms, aerodynamics, and control systems. Moreover, we expect that one can research FWV flight by CAD model import,joint connection, and parameter setup. Therefore, we establish a general and simple simulation platform framework based on MATLAB/Simulink with Simscape Multibody Toolbox for FWV mechanism, aerodynamics, and formation control for hovering or forward flight. Inspired by Purdue Hummingbird design, we provided simplified flapping wing aerodynamics with passive rotation, virtual environment interaction, adaptive cascade control, and consensus formation algorithms in the MATLAB integration platform. This work is fundamental for the verification of research, and useful for the improvement of mechanic design,actuator selection, and control algorithms of the ontology or swarm thanks to its compatibility.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Design and Optimization of Sensing-Communication Integration System in Industrial Automation Based on FDA and PA 25

Design and Optimization of Sensing-Communication Integration...

引用

25th IEEE International Conference on Industrial technology, ICIT 2024

作者： Feng, Freddy Y.P. Mi, De Guan, Mingxiang Sun, Lichao Tang, Hui Liu, Chang Xiao, Pei Lu, Chen Shenzhen Institute Of Information Technology Shenzhen China College Of Computing Birmingham City University Birmingham United Kingdom School Of Engineering Lancaster University Lancaster United Kingdom Extend Robotics Limited Reading United Kingdom Institute For Communication Systems University Of Surrey Guildford United Kingdom

ISBN: (纸本)9798350340266

Industrial automation is a rapidly evolving field that aims to improve efficiency, productivity, and safety in various verticals. The integration of sensing and communication technologies has become crucial in this automation process. Re-cent advancements in Frequency Diverse Array (FDA) antennas have shown promising performance advantages over traditional phased array antennas in certain aspects, e.g., the FDA-based radar can focus its beam in both angle and range domains simultaneously. In this paper, we propose a novel architecture for sensing-communication integration in the context of industrial automation, where FDA antennas are employed as the radar sensing array, while phased array antennas are utilized for communication purposes. The integration of FDA antennas brings benefits such as more accurate object detection, improved spatial resolution, and efficient data transmission. We present a comprehensive analysis of the proposed architecture, including design considerations, system modeling, and performance evalu-ation. The results demonstrate the potential of FDA antennas in enhancing the sensing-communication integration in industrial automation, leading to improved efficiency, productivity, and safety in industrial environments. © 2024 IEEE.

关键词： Integration

来源：评论

学校读者我要写书评

暂无评论

The Influence of Dynamic Game Difficulty Balancing on Player Experience in Puzzle Games 25

The Influence of Dynamic Game Difficulty Balancing on Player...

引用

25th IEEE International Conference of Young Professionals in Electron Devices and Materials, EDM 2024

作者： Khudaybergenov, Timur A. Turdiev, Temur T. Khafizov, Murad R. Kugurakova, Vlada V. Dept. Information Technology Urgench Branch of Tashkent University of Information Technologies named after Muhammad al-Khwarizmi Urgench Uzbekistan Dept. Information Security Urgench Branch of Tashkent University of Information Technologies named after Muhammad al-Khwarizmi Urgench Uzbekistan Institute of Information Technology and Intelligent Systems Kazan Federal University Kazan Russia

ISBN: (纸本)9798350389234

This research investigates the interplay between Dynamic Game Difficulty Balancing (DGDB) and Game User Experience (GUX) within the realm of puzzle gaming. The primary objective was to examine how DGDB influences user experience, with a specific focus on enhancing player engagement, deepening immersion, and facilitating the occurrence of "a-ha" moments. The study employed a controlled experimental approach, where data was collected and subsequently analyzed both quantitatively and qualitatively. This involved assessing players' objective in-game performance metrics alongside their subjective feedback gathered via post-game questionnaires. The results indicate that while DGDB can positively affect player engagement and overall satisfaction, the extent of this impact is relatively moderate, as indicated by improvements in game completion rates and the experience of flow states. The study also delved into the effects of DGDB on 'a-ha' moments, yielding mixed outcomes. It became evident through this research that the dynamic between DGDB and puzzle game design is intricate, heavily influenced by varying player preferences and perceptions. © 2024 IEEE.

关键词： Game design

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：