检索结果-内蒙古大学图书馆

ieee International Conference on Industrial Technology (ICIT)

作者： Xin Wen Zhen-Tao Liu Dan Chen School of Automation China University of Geosciences Wuhan China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education

ISBN: (数字)9798331521950

ISBN: (纸本)9798331521967

Facial expression recognition plays a key role in promoting the development of comprehensive intelligence and building friendly human-computer interaction. Due to the interference of feature noise in expression data, the lightweight facial expression recognition model with fewer parameters is difficult to learn more expression features through simple training, which limits the improvement of its recognition performance. An efficient facial expression recognition network based on Spot-adaptive Knowledge Distillation is proposed in this paper. Inspired by VoVNetV2, the network designed in this paper is lightweightly improved using Depthwise Separable Convolution and the parameter-free SimAM attention mechanism, reducing the number of parameters to 0.21 M. To further improve the recognition accuracy of the model, Spot-adaptive Knowledge Distillation is employed to improve the characterization ability of the model. The recognition accuracies of the student network designed in this paper on the KDEF and RAF-DB datasets are 93.05% and 81.17% respectively after spot-adaptive distillation.

关键词： Knowledge engineering Training Human computer interaction Accuracy Face recognition computational modeling Noise Neural networks Interference Data models

来源：评论

学校读者我要写书评

暂无评论

2009 ieee International symposium on computational intelligence in Robotics and automation, CIRA 2009

2009 IEEE International Symposium on Computational Intellige...

引用

2009 ieee International symposium on computational intelligence in Robotics and automation, CIRA 2009

The proceedings contain 99 papers. The topics discussed include: grasp motion planning for box opening task by multi-fingered hands and arms;planning of diverse complex cooperative robot actions using multi-stage genetic algorithm;exploration and mapping for unstructured robot teams;stable trajectory generator-echo state network trained by particle swarm optimization;modular reservoir computing networks for imitation learning of multiple robot behaviors;interactive evolutionary computation for robot design support system;CMAC neural networks structures;robot deception: recognizing when a robot should deceive;a small world algorithm for high-dimensional function optimization;stochastic approach on a simplified OCC model for uncertainty and believability;examining robotic systems with shape-adjustable manipulators under dynamic environments: from simulation to verification;and SILT: scale-invariant line transform.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Salp Swarm Algorithm Based Fractional Order PID controller Design for Trajectory Tracking Simulation of an Underactuated Underwater Vehicle Model

Salp Swarm Algorithm Based Fractional Order PID Controller D...

引用

International symposium on Underwater Technology

作者： Gülten Yılmaz Serhat Yılmaz Electronics and Automation Department Kocaeli University Hereke Asım Kocabıyık Vocational School Kocaeli Turkey Department of Electronics and Communication Kocaeli University Engineering Faculty Kocaeli Turkey

ISBN: (数字)9798331534189

ISBN: (纸本)9798331534196

Autonomous Underwater Vehicles (AUVs) are used in many fields, primarily in scientific, commercial, and military areas. These vehicles require various guidance and control systems to autonomously perform tasks underwater. The modeling, system identification, and control of AUVs remain active areas of research and development. In this study, modelling and trajectory tracking simulation of a low-speed, modular, symmetrical and underactuated underwater vehicle with 4DOF are evaluated. The vehicle, named RedFin is developed in Kocaeli University Laboratories. Dynamical model parameters of the AUV's are estimated by computational Fluid Dynamics (CFD) and analytical calculation techniques and the AUV mathematical model is obtained. Model validation was carried out through static and dynamic pool tests. control algorithm has been developed, offering more precise solutions. For dynamic models described by fractional differential equations, a Fractional Order PID (FOPID) control algorithm was developed, offering more precise solutions. In the study, a PID controller and a FOPID controller were designed for the trajectory tracking control of the AUV. The Salp Swarm Algorithm (SSA), one of the heuristic optimization algorithms, is used to optimize the gain parameters of both controllers. The integral of the absolute value of the position error (IAE) is selected as the fitness function. “Band Limited White Noise” with 0.5 W power and 200 Hz frequency was added to the system model control inputs. During the orbit tracking, the RMS values of depth control errors are very small, such as 0.021 m for SSA-PID, 0.0032 m for GA-KDPID, and 0.00118 m for SSA-KDPID.

关键词： Power demand Trajectory tracking Heuristic algorithms computational modeling computational fluid dynamics control systems Mathematical models Robustness Trajectory Vehicle dynamics

来源：评论

学校读者我要写书评

暂无评论

Exploiting the Intrinsic Neighborhood Semantic Structure for Domain Adaptation in EEG-based Emotion Recognition

引用

ieee Transactions on Affective Computing 2025年

作者： Yang, Yi Wang, Ze Song, Yu Jia, Ziyu Wang, Boyu Jung, Tzyy-Ping Wan, Feng Macau University of Science and Technology Macao Centre for Mathematical Sciences Respiratory Disease AI Laboratory on Epidemic Intelligence and Medical Big Data Instrument Applications Faculty of Innovation Engineering 999078 China Tianjin University of Technology School of Electrical Engineering and Automation Tianjin Key Laboratory of New Energy Power Conversion Transmission and Intelligent Control Tianjin300384 China Chinese Academy of Sciences Beijing Key Laboratory of Brainnetome and Brain-Computer Interface and Brainnetome Center Institute of Automation Beijing100045 China Western University Department of Computer Science Brain Mind Institute LondonONN6A 3K7 Canada University of California at San Diego Swartz Center for Computational Neuroscience Institute for Neural Computation La Jolla CA92093 United States University of Macau Department of Electrical and Computer Engineering Faculty of Science and Technology China University of Macau Centre for Cognitive and Brain Sciences Centre for Artificial Intelligence and Robotics Institute of Collaborative Innovation 999078 China

Due to the inherent non-stationarity and individual differences present in electroencephalogram (EEG) signals, developing a generalizable model that performs well on new subjects is challenging in EEG-based emotion recognition. Most existing domain adaptation (DA) methods typically mitigate these discrepancies by aligning the marginal distributions of domain feature representations. However, when there is a significant difference in the class-conditional distribution between domain features and labels, the domain-invariant features learned by aligning marginal distributions may have limited discriminative ability for unlabeled target instances or even prove counterproductive. To address this issue, we propose a Neighborhood Semantic Aware Learning-based Dynamic Graph Attention Convolution (NSAL-DGAT) approach that learns target semantic information by considering the inter-domain semantic topological structure, thereby improving classifier adaptation for target instances. Specifically, the proposed NSAL framework is designed to capitalize on the insight that after domain feature alignment, some target samples and their neighboring source samples exhibit similar semantics. By leveraging the neighborhood topological structure, we extract and incorporate semantic target features to train a more transferable classifier. Besides, we implement an entropy weighting mechanism to emphasize representative target semantic information, encouraging target instances to prioritize high-confidence individuals within the source neighborhood. We have conducted extensive experiments on the public SEED dataset and our collected the Hearing-Impaired EEG Dataset (HIED). The experimental results underscore the efficacy of our proposed NSAL-DGAT approach, showcasing state-of-the-art accuracy in subject-dependent as well as subject-independent scenarios. The source code is available at https://***/YYingDL/NSAL-DGAT. © 2010-2012 ieee.

关键词： Speech analysis

来源：评论

学校读者我要写书评

暂无评论

Collaborative Optimization Algorithm of Deep Learning and Compliant control in Online Trajectory Planning

Collaborative Optimization Algorithm of Deep Learning and Co...

引用

International Conference on Communication Systems and Network Technologies (CSNT)

作者： Junting Liao Chunjian Zhao Hui Xie School of Computer and Artificial Intelligence Xiangnan University Chenzhou China China Telecom Hunan Chenzhou Branch Chenzhou China

ISBN: (数字)9798331531935

ISBN: (纸本)9798331531942

With the rapid development of robotics and automation, online trajectory planning faces problems such as poor adaptability to dynamic environments and low computational efficiency. This study aims to propose a collaborative optimization algorithm that combines deep learning and compliant control to improve the efficiency and accuracy of online trajectory planning. First, CNN is used for feature extraction and a prediction model is established. Then a compliant control strategy is designed to deal with dynamic obstacles to ensure the flexibility and stability of the robot in trajectory planning. The trained CNN model is then used to generate real-time trajectory predictions. The robot motion is adjusted through a fuzzy logic controller to optimize the relative position of the trajectory and obstacles. The average obstacle avoidance success rate of the algorithm in dynamic environments reaches 97%. The collaborative optimization algorithm of deep learning and compliant control effectively improves the real-time and accuracy of online trajectory planning, providing a new solution for the application of intelligent robots in complex environments.

关键词： Deep learning Trajectory planning Heuristic algorithms Collaboration Predictive models Prediction algorithms Real-time systems Trajectory Convolutional neural networks Optimization

来源：评论

学校读者我要写书评

暂无评论

SwinLip: An Efficient Visual Speech Encoder for Lip Reading Using Swin Transformer

引用

Neurocomputing 2025年 639卷

作者： Park, Young-Hu Park, Rae-Hong Park, Hyung-Min Department of Artificial Intelligence Sogang University Seoul04107 Korea Republic of Department of Electronic Engineering Sogang University Seoul04107 Korea Republic of

This paper presents an efficient visual speech encoder for lip reading. While most recent lip reading studies have been based on the ResNet architecture and have achieved significant success, they are not sufficiently suitable for efficiently capturing lip reading features due to high computational complexity in modeling spatio-temporal information. Additionally, using a complex visual model not only increases the complexity of lip reading models but also induces delays in the overall network for multi-modal studies (e.g., audio–visual speech recognition, speech enhancement, and speech separation). To overcome the limitations of Convolutional Neural Network (CNN)-based models, we apply the hierarchical structure and window self-attention of the Swin Transformer to lip reading. We configure a new lightweight scale of the Swin Transformer suitable for processing lip reading data and present the SwinLip visual speech encoder, which efficiently reduces computational load by integrating modified Convolution-augmented Transformer (Conformer) temporal embeddings with conventional spatial embeddings in the hierarchical structure. Through extensive experiments, we have validated that our SwinLip successfully improves the performance and inference speed of the lip reading network when applied to various backbones for word and sentence recognition, reducing computational load. In particular, our SwinLip demonstrated robust performance in both English LRW and Mandarin LRW-1000 datasets and achieved state-of-the-art performance on the Mandarin LRW-1000 dataset with less computation compared to the existing state-of-the-art model. © 2025 Elsevier B.V.

关键词： Speech enhancement

来源：评论

学校读者我要写书评

暂无评论

Two Time Scale Partial Unknown Dynamics System Tracking control Based on Off-Policy Inverse Reinforcement Learning 11th

Two Time Scale Partial Unknown Dynamics System Tracking Cont...

引用

11th International symposium on computational intelligence and Industrial Applications, ISCIIA 2024

作者： Shen, MinYin Liu, Fei Institute of Automation Jiangnan University Wuxi214122 China

ISBN: (纸本)9789819647552

This article, integrating the singular perturbation technique with inverse reinforcement learning, proposes a novel linear two time scale system tracking control method grounded in off-policy inverse reinforcement learning. This method addresses the challenge of unknown cost functions prevalent in industrial processes. First, the singular perturbation method is leveraged to decompose the original problem into fast and slow subsystem issues. Without manually designing a cost function, the method learns from known optimal behavioral data by reconstructing cost functions tailored to each subsystem, enabling the system to mimic optimal behaviors. Then, for the fast time scale system, a model-based inverse reinforcement learning method is adopted, while for the slow time scale system, a model-free off-policy inverse reinforcement learning strategy is employed, which reconstructs the system’s cost function solely using measured expert behavioral data inputs. Finally, using a mixed separation thickening industrial process to illustrate the effectiveness of this method in two time scale tracking. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Cost functions

来源：评论

学校读者我要写书评

暂无评论

Expanding conformal prediction to system identification

引用

Pattern Recognition 2025年 167卷

作者： Msaddi, Saleh Kumbasar, Tufan Artificial Intelligence and Intelligent Systems Laboratory Istanbul Technical University Istanbul34469 Turkey Control and Automation Engineering Department Istanbul Technical University Istanbul34469 Turkey

System Identification (SysID) is a cornerstone of control engineering, aimed at modeling system dynamics from input–output data. However, SysID approaches often overlook uncertainty quantification (UQ), which is crucial for robust decision-making. To address this gap, we propose Conformal System Identification (C-SysID), a model-agnostic framework for UQ in single-output and multiple-output SysID tasks. C-SysID transforms point predictions of any SysID model into High-Quality (HQ) Prediction Intervals (PIs) for single-output systems or Prediction Regions (PRs) for multiple-output systems, ensuring the true output is captured within the specified confidence level. For single-output systems, we use the Updated Residuals (UR) and Weighted Updated Residuals (WUR) methods to generate valid HQ PIs. In the case of multiple-output systems, we apply the Copula method and introduce the Updated Copula (UC) method, which captures output dependencies to generate HQ PRs. We empirically validate C-SysID on diverse single-output and multiple-output SysID datasets. The statistically significant comparative results show that the WUR method outperforms baseline methods, delivering tight PIs with the desired coverage. For multiple-output systems, our proposed UC method generates PRs that achieve accurate joint coverage while maintaining the compactness of PRs. The results highlight the effectiveness of C-SysID in UQ and enhancing uncertainty-aware predictions for SysID applications. © 2025 Elsevier Ltd

关键词： Parameter estimation

来源：评论

学校读者我要写书评

暂无评论

Multi-objective Evolutionary Algorithm Based Graph Neural Network Architecture Search 11th

Multi-objective Evolutionary Algorithm Based Graph Neural Ne...

引用

11th International symposium on computational intelligence and Industrial Applications, ISCIIA 2024

作者： He, Lianyi Liu, Xiaobo Xiang, Hongbo Wang, Guangjun School of Automation China University of Geosciences Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan430074 China Key Laboratory of Geological Survey and Evaluation of Ministry of Education China University of Geosciences Wuhan430074 China

ISBN: (纸本)9789819647521

Graph Neural Networks (GNN) has become a powerful graph data processing method, which has been widely used in node classification, link prediction, and other graph analysis tasks. Due to the diversity and complexity of graph structures and information propagation, as well as the handling of heterogeneous graphs, etc., the design of GNNs presents many challenges. The existing methods have high classification accuracy, but their structures are very complex. In this paper, a graph neural network architecture search framework named MO-GNN, which is based on multi-objective evolutionary algorithm is proposed. Furthermore, a weight sharing strategy MO-GNN-WS is proposed to reduce the resource consumption caused by weight training of different architectures. To verify the performance of the proposed algorithm, experiments on four popular graph datasets are used for transduction and induction tasks. The experimental results show that the MO-GNN-WS algorithm outperforms the most advanced neural network architecture search methods in terms of classification accuracy and resource consumption. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Data Sharing

来源：评论

学校读者我要写书评

暂无评论

Proceedings 2003 ieee International symposium on computational intelligence in Robotics and automation. computational intelligence in Robotics and automation for the New Millennium (Cat. No.03EX694)

Proceedings 2003 IEEE International Symposium on Computation...

引用

ieee International symposium on computational intelligence in Robotics and automation (CIRA)

The following topics are dealt with: reinforcement learning; neural networks and its applications; tele-existence; fuzzy control for robots; robot assisted therapy and activity; pet-type robot for aged people's welfare; rehabilitation support system; human-robot interaction; emergence of adaptive behavior, behavior based architecture; new trends in robotics; hybrid intelligent systems in robotics; co-creation system; cooperative mobile robots; evolutionary computation; robot design; micro robot; computational intelligence in industrial applications; vision-based robotics; virtual reality; navigation system; artificial intelligence and application; fuzzy pattern recognition; intelligent systems and nonlinear control; dynamic path planning; collision detection; evolutionary robotics; modular robotics; virtual environment; global localization; pattern recognition by neural networks and evolutionary algorithms; digital human; social learning; biologically inspired robotics; manipulator analysis; manipulator control; learning from agent, human, and world; distributed autonomous systems; future manufacturing; bio-mimetric robotics; computer vision and pattern recognition; fuzzy neural network for modeling; fuzzy neural network for control; adaptive control; agent-based socio-economic simulation; intelligent method for industrial and robotics application; multi-agent systems; advanced fuzzy control; image recognition and reconstruction; sensor fusion.

关键词： Robots Intelligent control control engineering

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：