检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Sun, Zhengnan Shi, Zhaotai Chen, Jiayin Liu, Qingtao Cui, Yu Ye, Qi Chen, Jiming College of Control Science and Engineering Zhejiang University Hangzhou310027 China State Key Laboratory of Industrial Control Technology Zhejiang University China Key Key Lab of CS&AUS of Zhejiang Province China

Bimanual dexterous manipulation remains significant challenges in robotics due to the high DoFs of each hand and their coordination. Existing single-hand manipulation techniques often leverage human demonstrations to guide RL methods but fail to generalize to complex bimanual tasks involving multiple sub-skills. In this paper, we introduce VTAOBiManip, a novel framework that combines visual-tactile-action pretraining with object understanding to facilitate curriculum RL to enable human-like bimanual manipulation. We improve prior learning by incorporating hand motion data, providing more effective guidance for dual-hand coordination than binary tactile feedback. Our pretraining model predicts future actions as well as object pose and size using masked multimodal inputs, facilitating cross-modal regularization. To address the multiskill learning challenge, we introduce a two-stage curriculum RL approach to stabilize training. We evaluate our method on a bottle-cap unscrewing task, demonstrating its effectiveness in both simulated and real-world environments. Our approach achieves a success rate that surpasses existing visual-tactile pretraining methods by over 20%. Copyright © 2025, The Authors. All rights reserved.

关键词： Bottles

来源：评论

学校读者我要写书评

暂无评论

EAVFormer: an end-to-end audio and visual emotion recognition network based on transformers

引用

Multimedia Systems 2025年第3期31卷 1-22页

作者： Liu, Haoran Sun, Zijian Li, Haibin Li, Yaqian Zhang, Wenming Song, Tao School of Information Science and Engineering Yanshan University Qinhuangdao066004 China The Key Laboratory for Special Fiber and Fiber Sensor of Hebei Province Yanshan University Qinhuangdao066004 China Key Lab of Industrial Computer Control Engineering of Hebei Province Yanshan University Qinhuangdao066004 China Shenzhen Research Institute of Yanshan University Shenzhen518057 China Key Lab of Measurement Technology & Instrumentation of Hebei Province School of Electrical Engineering Yanshan University Qinhuangdao066004 China

Audio-visual emotion recognition, integrating information from various sources, enhances the comprehensive understanding and analysis of human emotions. However, existing methods in this field primarily focus on feature fusion, often overlooking the equally crucial stage of feature extraction. Typically, these methods employ pre-trained models to extract features during the initial phase, complicating modality-specific feature extraction and increasing system complexity and coordination challenges, which hinder effective audio-visual feature fusion. To address these issues, we propose an end-to-end audio-visual emotion recognition model, EAVFormer. For the audio modality, we input raw audio signals along with a mix of mel spectrograms and Mel Frequency Cepstral Coefficients (MFCCs), employing a hierarchical transformer technique that facilitates both shallow local and deep global feature extraction to fully harness complementary acoustic information, significantly enhancing the acoustic modality’s contribution. For the visual modality, we input raw video frames and employ a hierarchical transformer feature extraction method that integrates three-dimensional convolution and spatio-temporal self-attention mechanisms to address spatial and temporal redundancies and dependencies, thereby enhancing the expression of dynamic visual information. In the modal fusion stage, we construct dual-modality representations and encode inter-modal information while preserving modality-specific and meaningful intra-modal features. Testing on three popular datasets RAVDESS, CREMA-D, and CMU-MOSEI demonstrates that EAVFormer surpasses existing state-of-the-art methods in audio-visual emotion recognition accuracy, achieving rates of 95.17%, 86.13%, and 78.22%, respectively. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Image coding

来源：评论

学校读者我要写书评

暂无评论

Contextual Multi-Armed Bandit-Based Dynamic Cooperative Link Configuration for AUV in UWASNs With Energy Harvesting

引用

IEEE Transactions on Cognitive Communications and Networking 2025年

作者： Dai, Jun Li, Xinbin Han, Song Yu, Junzhi Liu, Zhixin Yanshan University Key Lab of Industrial Computer Control Engineering of Hebei Province Qinhuangdao066004 China Yanshan University Key Laboratory of Intelligent Rehabilitation and Neuromodulation of Hebei Province Qinhuangdao066004 China Peking University State Key Laboratory for Turbulence and Complex Systems Department of Advanced Manufacturing and Robotics BIC-ESAT College of Engineering Beijing100871 China

This paper investigates the cooperative link configuration problem for Autonomous Underwater Vehicle (AUV) in Underwater Acoustic (UWA) sensor networks with Energy Harvesting (EH), which aims to maximize long-term cumulative capacity by jointly optimizing cooperation relay, AUV transmission power, and relay transmission power. Subject to unknown time-varying Channel State Information (CSI), unpredictable stochastic EH, and variable communication topology caused by AUV mobility, the proposed problem is an unknown dynamic combination decision optimization problem. To resist the change of the dynamic UWA communication topology with unknown time-varying CSI and EH, a novel collaborative contextual multi-armed bandit learning framework is proposed, which allows relay nodes to learn cooperatively and enriches the learning information of relays. Consequently, the proposed learning framework can resist the change of the dynamic UWA communication topology efficiently, thereby achieving the superior link configuration quickly. Besides, a learning-rate adjustment rule for the dynamic UWA communication topology is proposed to adaptively balance the exploration-exploitation, thereby avoiding missing the real superior joint relay-power configuration. Finally, simulation results show the significantly superiority of the proposed scheme. © 2015 IEEE.

关键词： Cooperative communication

来源：评论

学校读者我要写书评

暂无评论

Gesture Recognition of sEMG Based on Res-LSTM 17th

Gesture Recognition of sEMG Based on Res-LSTM

引用

17th International Conference on Intelligent Robotics and Applications, ICIRA 2024

作者： Zhao, Yujia Zou, Chunlong Yun, Juntong Jiang, Du Huang, Li Liu, Ying Jiang, Guozhang Xie, Yuanmin Key Laboratory of Metallurgical Equipment and Control Technology of Ministry of Education Wuhan University of Science and Technology Wuhan430081 China College of Mechanical Engineering Hubei University of Automotive Technology Shiyan442000 China Research Center for Biomimetic Robot and Intelligent Measurement and Control Wuhan University of Science and Technology Wuhan430081 China Hubei Key Laboratory of Mechanical Transmission and Manufacturing Engineering Wuhan University of Science and Technology Wuhan430081 China Hubei Province Key Laboratory of Intelligent Information Processing and Real-Time Industrial System Wuhan University of Science and Technology Wuhan430081 China College of Computer Science and Technology Wuhan University of Science and Technology Wuhan430081 China School of Mechanical Engineering Hubei Engineering University Xiaogan432000 China

ISBN: (纸本)9789819607761

sEMG (surface electromyography) signal control of bionic prostheses has been widely studied over the past few years. In particular, sparse sEMG signals are rapidly developing in the field of gesture recognition for their convenience, noninvasiveness, and ease of access. However, compared with high-density EMG signals, sparse EMG signals lack rich feature information, which in turn affects gesture recognition accuracy. In order to reduce the loss of feature information of sparse EMG signals in the spatio-temporal dimension, this paper proposes a hybrid neural network Res-LSTM combining residual network and long short-term memory network. The ordinary convolutional blocks in the CNN network are replaced with residual blocks, and the final fully connected layer is removed and a constant mapping layer is added to adequate extraction of spatial feature information of the data. The Res layer’s output is utilized as the input for the long and short-term memory (LSTM) network, which further extracts the features of the data in the temporal dimension, and finally completes the categorization output through a fully connected layer. The average accuracy of 91.11% for gesture-motion classification was verified by training on a homebrew dataset;and tested on a public dataset, NinaPro DB1, with an average accuracy of 91.03%. The experimental findings indicate that the proposed Res-LSTM network framework contributes to solving the problem of lack of feature information of sparse sEMG signals and improves the pattern recognition accuracy. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Gesture recognition

来源：评论

学校读者我要写书评

暂无评论

Knowledge-driven inverse diffusion prediction algorithm for flexible job shop scheduling problem considering transportation resources and multiple breakdowns

引用

Swarm and Evolutionary Computation 2025年 96卷

作者： Wang, Cong Wei, Lixin Sun, Hao Hu, Ziyu Hebei Key Laboratory of Industrial Computer Control Engineering Yanshan University Qinhuangdao Hebei066004 China School of Electrical Engineering Yanshan University Qinhuangdao Hebei066004 China

The complex coupling between transportation and production poses significant challenges for scheduling algorithms, particularly in dynamic environments where contingencies are frequent. The traditional Flexible Job Shop Scheduling Problem (FJSP) does not fully account for these factors. Therefore, a two-stage knowledge-driven inverse diffusion prediction algorithm (TKIDP) is proposed for FJSP with transportation resources and multiple breakdowns (MBFJSPT). In the first phase, TKIDP operates in a static environment and aims to minimize the makespan and load imbalance, thereby reducing the risk of machine breakdowns. When a machine breakdown occurs, TKIDP enters a rescheduling phase and minimizes total energy consumption as an additional objective. An inverse diffusion prediction strategy is proposed to address the occurrence of multiple machine breakdowns. By learning from optimization processes of historical breakdowns, the proposed strategy identifies potential evolutionary patterns and predicts the evolutionary trends of future breakdowns. A similarity measure based on mutual entropy is introduced to select the most relevant historical breakdowns for learning. This method enhances prediction accuracy. To further improve exploration and exploitation capabilities, an adaptive competitive reconfiguration strategy is proposed. This strategy incorporates multiple deletion and reconstruction operators to guide the algorithm toward the optimal solution. To verify the effectiveness of MBFJSPT, a dedicated MB test set is constructed. On the MB1-18 test, TKIDP is compared with five other algorithms. Based on the Inverted generational distance (IGD), hypervolume (HV), and relative deviation (RD) metrics, TKIDP achieves the best results on 15, 18, and 18 out of the 18 test instances, respectively. © 2025 Elsevier B.V.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Language-Queried Target Sound Extraction Without Parallel Training Data

Language-Queried Target Sound Extraction Without Parallel Tr...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Ma, Hao Peng, Zhiyuan Li, Xu Li, Yukai Shao, Mingjie Kong, Qiuqiang Liu, Ju School of Information Science and Engineering Shandong University Qingdao China Department of Computer Science North Carolina State University NC United States ARC Lab Tencent PCG China Key Laboratory of System and Control AMSS Chinese Academy of Sciences Beijing China The Chinese University of Hong Kong Hong Kong

ISBN: (纸本)9798350368741

Language-queried target sound extraction (TSE) aims to extract specific sounds from mixtures based on language queries. Traditional fully-supervised training schemes require extensively annotated parallel audio-text data, which are labor-intensive. We introduce a parallel-data-free training scheme, requiring only unlabelled audio clips for TSE model training by utilizing the contrastive language-audio pre-trained model (CLAP). In a vanilla parallel-data-free training stage, target audio is encoded using the pre-trained CLAP audio encoder to form a condition embedding, while during testing, user language queries are encoded by CLAP text encoder as the condition embedding. This vanilla approach assumes perfect alignment between text and audio embeddings, which is unrealistic. Two major challenges arise from training-testing mismatch: the persistent modality gap between text and audio and the risk of overfitting due to the exposure of rich acoustic details in target audio embedding during training. To address this, we propose a retrieval-augmented strategy. Specifically, we create an embedding cache using audio captions generated by a large language model (LLM). During training, target audio embeddings retrieve text embeddings from this cache to use as condition embeddings, ensuring consistent modalities between training and testing and eliminating information leakage. Extensive experiment results show that our retrieval-augmented approach achieves consistent and notable performance improvements over existing state-of-the-art with better generalizability. © 2025 IEEE.

关键词： contrastive language-audio pre-training retrieval-augmented strategy target sound extraction uni-modal training for multi-modal tasks

来源：评论

学校读者我要写书评

暂无评论

SIMSE: a contrastive learning method combining sample importance metric and semantic enhancement

引用

Expert Systems with Applications 2025年 286卷

作者： Gao, Yangyang Zheng, Zhi Huang, Wenjun Lin, Xiaomin College of Computer and Cyber Security Fujian Normal University Fuzhou350117 China College of Control Science and Engineering Zhejiang University Hangzhou310027 China State Key Laboratory of Industrial Control Technology Zhejiang University Hangzhou310027 China

In complex application scenarios, the lack of accurate labels for training samples has positioned contrastive learning a key focus in self-supervised learning. This approach effectively extracts meaningful representations from unlabeled data. A significant challenge lies in identifying high-value samples and exploring their semantic features to improve performance. Most existing self-supervised learning methods do not focus much on mining or augmenting valuable samples, relying mainly on contrastive learning between original samples. As a result, performance suffers in cases with limited or sparse data. To address this, we propose a contrastive learning method that combines sample importance measurement and semantic enhancement, overcoming the limitations of traditional methods that only use original samples. First, we generate additional valuable samples using interpolation and apply a similarity-based strategy to better distinguish positive and negative samples, refining the sample partitioning process. Second, we design a semantic enhancement mechanism to better capture and strengthen shared high-level semantic features between samples. Third, we introduce a new metric to evaluate sample value by measuring oscillations in the learning model caused by gradients, determining the importance of each sample. We also use confidence learning to identify and correct mislabeled samples. Extensive evaluations conduct on multiple benchmark datasets demonstrate that our method improves linear classification accuracy by 2.23 %, 4.3 %, 1.6 %, 3.73 %, and 4.69 % on ImageNet-100, ImageNet-10, CIFAR-10, CIFAR-100, and STL-10, respectively. Additionally, accelerates convergence speed by 1.5x and effectively detects mislabeled samples. © 2025 Elsevier Ltd

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Physics-informed CGAN and multi-scale attention CNN for pipeline leakage diagnosis under imbalanced data

引用

Advanced engineering Informatics 2025年 66卷

作者： Zhu, Yongqiang Li, Shuaiyong Lang, Xianming Liu, Liang School of Computer Science and Technology Chongqing University of Posts and Telecommunications Chongqing400065 China Key Laboratory of Industrial Internet of Things and Networked Control Ministry of Education Chongqing University of Posts and Telecommunications Chongqing400065 China School of Information and Control Engineering Liaoning Petrochemical University Fushun113001 China

In industrial pipeline condition monitoring, class-imbalanced data distributions and multi-physics coupling effects significantly constrain the engineering applicability of intelligent diagnostic models. Class-imbalanced data lead to a scarcity of fault samples, making model training challenging. Meanwhile, complex multi-physics coupling effects complicate the prediction of equipment operating conditions. Generating synthetic data that are consistent in both statistical distribution and physical mechanism has become an effective solution to these issues. In this article, a hybrid framework that integrates a physics-informed data augmentation method with an intelligent fault diagnosis model is proposed. Within this framework, a novel physics-informed conditional generative adversarial network (PI-CGAN) is developed, where a physics-informed network (PIN) converts equipment behavior equations into implicit constraints for the generator, ensuring dual consistency in both statistical distributions and physical mechanisms. Subsequently, a hybrid attention module (HAM) is proposed to enhance data diversity and quality through cross-channel feature correlation and critical temporal segment localization. Furthermore, a multi-scale attention convolutional neural network (MSACNN) is developed to effectively decouple coupled features under complex operating conditions via a multi-scale attention mechanism. Experimental results demonstrate that under extreme data imbalance conditions, the proposed hybrid diagnostic framework can generate industrially viable synthetic data while achieving accurate fault identification. © 2025

关键词： Leak detection

来源：评论

学校读者我要写书评

暂无评论

Fuzzy adaptive robust control for space robot considering the effect of the gravity

引用

Chinese Journal of Aeronautics 2014年第6期27卷 1562-1570页

作者： Qin Li Liu Fucai Liang Lihuan Gao Jingfang Key Lab of Industrial Computer Control Engineering of Hebei Province Yanshan University

Space robot is assembled and tested in gravity environment, and completes on-orbit service（OOS） in microgravity environment. The kinematic and dynamic characteristic of the robot will change with the variations of gravity in different working condition. Fully considering the change of kinematic and dynamic models caused by the change of gravity environment, a fuzzy adaptive robust control（FARC） strategy which is adaptive to these model variations is put forward for trajectory tracking control of space robot. A fuzzy algorithm is employed to approximate the nonlinear uncertainties in the model, adaptive laws of the parameters are constructed, and the approximation error is compensated by using a robust control algorithm. The stability of the control system is guaranteed based on the Lyapunov theory and the trajectory tracking control simulation is performed. The simulation results are compared with the proportional plus derivative（PD） controller, and the effectiveness to achieve better trajectory tracking performance under different gravity environment without changing the control parameters and the advantage of the proposed controller are verified.

关键词： Fuzzy adaptive Microgravity Robustness Space robot Trajectory tracking control

来源：评论

学校读者我要写书评

暂无评论

Collision avoidance for a mobile robot based on radial basis function hybrid force control technique

引用

Chinese Physics B 2009年第10期18卷 4222-4228页

作者：温淑焕 Key Lab of Industrial Computer Control Engineering of Hebei Province Yanshan University

Collision avoidance is always difficult in the planning path for a mobile robot. In this paper, the virtual force field between a mobile robot and an obstacle is formed and regulated to maintain a desired distance by hybrid force control algorithm. Since uncertainties from robot dynamics and obstacle degrade the performance of a collision avoidance task, intelligent control is used to compensate for the uncertainties. A radial basis function （RBF） neural network is used to regulate the force field of an accurate distance between a robot and an obstacle in this paper and then simulation studies are conducted to confirm that the proposed algorithm is effective.

关键词： mobile robot collision avoidance hybrid force/position control path planning RBF neural network

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：