检索结果-内蒙古大学图书馆

2025 IEEE International Conference on mechatronics, ICM 2025

作者： Choi, Wonhyeok Oh, Sehoon DGIST Department of Robotics and Mechatronics Engineering Daegu Korea Republic of

ISBN: (纸本)9798331533892

This paper proposes a generalized dynamics model for a Non-Grounded System (NGS) mounted with a Reaction Force Sensing Series Elastic Actuator (RFSEA), focusing on scenarios where the actuator's base is unfixed and interacts with the ground. In RFSEA-based force control, the actuator base, which provides the necessary reaction force as dictated by Newton's third law, plays a critical role. This study examines how the interaction between the actuator base and the ground it is attached to influences the response of the series elasticity that generates force in the RFSEA. The investigation involves two steps. First, a generalized dynamics model that considers the dynamics between the actuator base and the ground is derived. Second, the Small Gain Theorem (SGT) is applied to evaluate the robustness of force control under load-side model uncertainties when a Force-based Disturbance Observer (FDOB), originally designed for GS, is adapted to NGS. The analysis reveals that force control utilizing FDOB can become unstable when the actuator base is unfixed, and simulations validate these findings. © 2025 IEEE.

关键词： Electric grounding

来源：评论

学校读者我要写书评

暂无评论

An Integrated Setup for Sequential Hybrid Laser Beam Micromachining (LBMM) and Micro-EDM for Machining Micro-hole Arrays with Automated Tool Localization

引用

International Journal of Precision engineering and Manufacturing 2025年 1-18页

作者： Azhar, Wan Ahmad Bin Wan Saleh, Tanveer Autonomous Systems and Robotics Research Unit Department of Mechatronics Engineering International Islamic University Malaysia Jalan Gombak Kuala Lumpur53100 Malaysia

Micro holes array has significant applications in the field of aerospace and turbo machinery, industrial filtration, microfluidic devices, biomedical engineering, and so on. There are various ways to manufacture micro holes arrays. In order to have a cost-effective, yet precise and efficient process for the fabrication of micro holes array, an integrated setup has been introduced in this paper to combine Laser Beam Micro Machining (LBMM) and micro-EDM (μEDM). This approach uses LBMM for quick material removal and µEDM for precise finishing. However, one of the main challenges in hybrid LBMM-µEDM is the localization of LBMM micro holes for accurate positioning of μEDM electrode. In this present study, an integrated setup with automated hole localization has been presented to improve the hybrid LBMM-μEDM process in terms of alignment accuracy and machining rate. The microscope camera that is mounted on the μEDM stage scans and captures images of LBMM holes. Then, the images were stitched and processed using image processing techniques to get the center coordinate of the LBMM micro holes. Finally, the operator validated the coordinates and G-code was automatically generated for the µEDM process to perform the fine finishing. The result indicated that the alignment accuracy is within 15 μm. Furthermore, the newly developed automation strategies for the hybrid LBMM-μEDM process show significant improvement in machining time with 5 × faster than the pure μEDM process, and better performances in overcut, taper angle, and electrode wear. Furthermore, the hybrid process was found to be 1.6 × more energy efficient as compared to the pure µEDM process. Overall, the newly developed integrated system was found to be highly effective for producing micro holes array. © The Author(s), under exclusive licence to Korean Society for Precision engineering 2025.

关键词： Precision engineering

来源：评论

学校读者我要写书评

暂无评论

AREPO: Uncertainty-Aware Robot Ensemble Learning under Extreme Partial Observability

引用

IEEE robotics and Automation Letters 2025年第6期10卷 5737-5744页

作者： Du, Yurui Hanut, Louis Bruyninckx, Herman Detry, Renaud Research Unit Processing Speech and Images KU Leuven Department of Electrical Engineering LeuvenB-3000 Belgium Research Unit Robotics Automation and Mechatronics KU Leuven Department of Mechanical Engineering LeuvenB-3000 Belgium

Real-world applications of vision-based robot learning face two major challenges: extreme partial observability and effective simulation-to-reality (sim-to-real) transfer. This letter introduces a robust robot learning framework that enhances uncertainty awareness to address these challenges. We reinterpret variational-autoencoder-based visual reinforcement learning (RL) from an uncertainty-quantification perspective, enabling resilience to high sensory noise and severe visual occlusions-common in industrial robotic tasks. To further improve sim-to-real transfer, we propose an uncertainty-aware ensemble RL algorithm. We validate our methods on a laboratory task designed as a proxy for real-world industrial applications characterized by harsh environments with low visibility and physical occlusions. Both simulation and real-world results demonstrate significant improvements in task accuracy and efficiency over various baselines, highlighting the benefits of uncertainty-aware robot learning for complex operational contexts. © 2016 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

A Study on Improving the Performance of Dynamic Model Based Redundant Navigation System for Sensor Fault Scenarios

引用

Journal of Institute of Control, robotics and Systems 2025年第4期31卷 351-359页

作者： Ko, Bo Sung Choi, Hyeong Jun Jeon, Myung Jun Cho, Hyeon Jin Song, Jin Woo Department of Artificial Intelligence and Robotics and Convergence Engineering for Intelligent Drone AIRI Sejong University Korea Republic of Department of Intelligent Mechatronics Engineering and Convergence Engineering for Intelligent Drone AIRI Sejong University Korea Republic of Agency for Defense Development Maritime Technology Research Institute Korea Republic of

This study proposes a redundant navigation system for UUV (Unmanned Underwater Vehicle) utilizing a 6 DOF (Degree of Freedom) VDM (Vehicle Dynamic Model). The system is designed to enable safe UUV asset return when all navigation sensors become inoperable. Since it is impossible to perfectly determine VDM coefficients through water tank experiments, a method to improve VDM output accuracy in real time during UUV mission is necessary. Here in this study, we introduce real-time adaptive 6 DOF VDM output error compensation algorithm to enhance the performance of redundant navigation systems. This research aims to establish the foundation for backup navigation solutions in situations when critical UUV sensor failures occur. © ICROS 2025.

关键词： Water tanks

来源：评论

学校读者我要写书评

暂无评论

Multiparty Random Phase Wrapping Secret-Sharing Systems for Visual Data Security

引用

IEEE Transactions on Systems, Man, and Cybernetics: Systems 2025年第5期55卷 3586-3600页

作者： Kim, Youhyun Jeong, Ongee Moon, Inkyu Javidi, Bahram Daegu Gyeongbuk Institute of Science and Technology Department of Robotics and Mechatronics Engineering Daegu42988 Korea Republic of AhnLab Inc. Gyeonggi Seongnam13493 Korea Republic of University of Connecticut Department of Electrical and Computer Engineering StorrsCT06269 United States

A secret sharing scheme is an important cryptographic procedure that enables the secure distribution of secret information, such as private images, in an untrusted network. However, in all secret sharing schemes, the sizes of the shares increase in proportion to the size of the secret information, because they involve computationally expensive polynomials of degree n-1 (n is total number of shares) and increasingly large modulus of modulo operations for security. Moreover, with large share sizes, it is not easy in practice to encrypt them quickly using block cipher algorithms. Therefore, ensuring the security of secret sharing for large-scale visual data with a reasonable share length and efficiently encrypting n shares properly represents formidable challenges. To overcome these challenges in secret sharing schemes, we propose new multiparty random phase wrapping secret sharing systems for visual datasets. Computational optical imaging enables the acquisition of wrapped phase information, ranging from–π to π, to be expressed in the form of a complex sinusoidal waveform. The proposed scheme allows for large-scale secret data—such as confidential digital images and visual data, including optical images—to be securely and efficiently shared and distributed to multiple parties or agencies utilizing a digital representation of a complex sinusoidal waveform of the secret information. The proposed scheme can be useful in cryptographic key escrow systems and in use with large-scale secret data. © 2013 IEEE.

关键词： Data Sharing

来源：评论

学校读者我要写书评

暂无评论

Communication Efficient Federated Learning for Multi-Organ Segmentation via Knowledge Distillation With Image Synthesis

引用

IEEE Transactions on Medical Imaging 2025年第5期44卷 2079-2092页

作者： Kim, Soopil Park, Heejung Chikontwe, Philip Kang, Myeongkyun Hwan Jin, Kyong Adeli, Ehsan Pohl, Kilian M. Hyun Park, Sang Daegu Gyeongbuk Institute of Science and Technology Department of Robotics and Mechatronics Engineering Daegu42988 Korea Republic of Korea University School of Electrical Engineering Seoul02841 Korea Republic of Stanford University Department of Psychiatry and Behavioral Sciences StanfordCA94305 United States

Federated learning (FL) methods for multi-organ segmentation in CT scans are gaining popularity, but generally require numerous rounds of parameter exchange between a central server and clients. This repetitive sharing of parameters between server and clients may not be practical due to the varying network infrastructures of clients and the large transmission of data. Further increasing repetitive sharing results from data heterogeneity among clients, i.e., clients may differ with respect to the type of data they share. For example, they might provide label maps of different organs (i.e. partial labels) as segmentations of all organs shown in the CT are not part of their clinical protocol. To this end, we propose an efficient communication approach for FL with partial labels. Specifically, parameters of local models are transmitted once to a central server and the global model is trained via knowledge distillation (KD) of the local models. While one can make use of unlabeled public data as inputs for KD, the model accuracy is often limited due to distribution shifts between local and public datasets. Herein, we propose to generate synthetic images from clients’ models as additional inputs to mitigate data shifts between public and local data. In addition, our proposed method offers flexibility for additional finetuning through several rounds of communication using existing FL algorithms, leading to enhanced performance. Extensive evaluation on public datasets in few communication FL scenario reveals that our approach substantially improves over state-of-the-art methods. © 1982-2012 IEEE.

关键词： Image segmentation Servers Data models Accuracy Computed tomography Computational modeling Federated learning Predictive models Decoding Training

来源：评论

学校读者我要写书评

暂无评论

A deep learning-based model for automated STN localization using local field potentials in Parkinson’s disease

引用

Soft Computing 2025年第4期29卷 2343-2362页

作者： Hosny, Mohamed Naeem, Mohamed A. Zhu, Minwei Gao, Wenpeng Elshenhab, Ahmed M. Fu, Yili Department of Electrical Engineering Benha Faculty of Engineering Benha University Benha Egypt State Key Laboratory of Robotics and Systems Harbin Institute of Technology Harbin150080 China Department of Mechatronics and Robotics Engineering Faculty of Engineering South Valley University Qena Egypt First Affiliated Hospital Harbin Medical University Harbin150001 China School of Life Science and Technology Harbin Institute of Technology Harbin150080 China Department of Mathematics Faculty of Science Mansoura University Mansoura35516 Egypt

Accurate identification of the subthalamic nucleus (STN) borders is time-consuming, relying heavily on the neurosurgeon expertise in manually interpreting the electrophysiological signals. Local field potentials (LFPs) have garnered imperative attention due to their strong correlation with the STN. However, existing detection models often face challenges with high computational complexity, hyperparameter optimization and lack of explainability, making them unreliable for clinicians. Therefore, this study introduces an explanatory framework using convolutional neural networks (CNN) for detecting the STN region from LFPs. Continuous wavelet transform is employed to convert LFPs signals into scalogram images, which are then processed by sixteen CNN models. We evaluated our framework by examining the impact of various limiting factors on the classification performance, including model size, learning rate (LR), optimizers and data scaling. Deep features are extracted from the top-performing CNN architectures to capture rich representations of the scalograms. These features are then fused and classified using k-nearest neighbour algorithm. Gradient-weighted class activation mapping is used to explain the decisions made by the proposed model. Our approach achieved an accuracy of 99.61%, outperforming individual CNN models for STN localization. The experimental results revealed that CNN models, embedded with additional hyperparameters and layers, generally outperformed smaller models. Besides, low LR significantly enhanced the performance compared to high LR. Moreover, features extracted from untuned networks produced lower performance than tuned networks. The proposed system could revolutionize deep brain stimulation surgery by increasing efficiency and reducing reliance on clinician expertise for STN detection. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.

关键词： Wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

Generalized Dynamics of Non-Grounded Systems with Series Elastic Actuators and Effect of Actuator Base Fixation on Force Control Robustness

Generalized Dynamics of Non-Grounded Systems with Series Ela...

引用

International Conference on mechatronics (ICM)

作者： Wonhyeok Choi Sehoon Oh Department of Robotics and Mechatronics Engineering DGIST Daegu Korea

ISBN: (数字)9798331533892

ISBN: (纸本)9798331533908

关键词： Actuators Analytical models Adaptation models Uncertainty Dynamics Force Elasticity Robustness Force control Load modeling

来源：评论

学校读者我要写书评

暂无评论

TAGA: A Tangent-Based Reactive Approach for Socially Compliant Robot Navigation Around Human Groups

arXiv

引用

arXiv 2025年

作者： Roy, Utsha Kumar Rahman, Sejuti The Department of Robotics and Mechatronics Engineering University of Dhaka Dhaka Bangladesh

Robot navigation in densely populated environments presents significant challenges, particularly regarding the interplay between individual and group dynamics. Current navigation models predominantly address interactions with individual pedestrians while failing to account for human groups that naturally form in real-world settings. Conversely, the limited models implementing group-aware navigation typically prioritize group dynamics at the expense of individual interactions, both of which are essential for socially appropriate navigation. This research extends an existing simulation framework to incorporate both individual pedestrians and human groups. We present Tangent Action for Group Avoidance (TAGA), a modular reactive mechanism that can be integrated with existing navigation frameworks to enhance their group-awareness capabilities. TAGA dynamically modifies robot trajectories using tangent action-based avoidance strategies while preserving the underlying model’s capacity to navigate around individuals. Additionally, we introduce Group Collision Rate (GCR), a novel metric to quantitatively assess how effectively robots maintain group integrity during navigation. Through comprehensive simulation-based benchmarking, we demonstrate that integrating TAGA with state-of-the-art navigation models (ORCA, Social Force, DS-RNN, and AG-RL) reduces group intrusions by 45.7–78.6% while maintaining comparable success rates and navigation efficiency. Future work will focus on real-world implementation and validation of this approach. Additional details and resources are available at https:// ***/***/taga/home. Copyright © 2025, The Authors. All rights reserved.

关键词： Population dynamics

来源：评论

学校读者我要写书评

暂无评论

ZAS-F: a Zero-shot Abstract Sub-goal Framework Empowers Robots for Long Horizontal Inventory Tasks

引用

IEEE Access 2025年 13卷 76400-76411页

作者： Wu, Yongshuai Zhang, Jian Wu, Shaoen Mao, Shiwen Wang, Ying Department of Information Technology Kennesaw State University Marietta 30060 GA United States Department of Electrical and Computer Engineering Auburn University Auburn 36849-5201 AL United States Department of Robotics and Mechatronics Engineering Kennesaw State University Marietta 30060 GA United States

Long-horizon tasks in unstructured environments are notoriously challenging for robots because they require the prediction of extensive action plans with thousands of steps while adapting to ever-changing conditions by reasoning among multimodal sensing spaces. Humans can efficiently tackle such compound problems by breaking them down into easily reachable abstract sub-goals, significantly reducing complexity. Inspired by this ability, we explore how we can enable robots to acquire sub-goal formulation skills for long-horizon tasks and generalize them to novel situations and environments. To address these challenges, we propose the Zero-shot Abstract Sub-goal Framework (ZAS-F), which empowers robots to decompose overarching action plans into transferable abstract sub-goals, thereby providing zero-shot capability in new task conditions. ZAS-F is an imitation-learning-based method that efficiently learns a task policy from a few demonstrations. The learned policy extracts abstract features from multimodal and extensive temporal observations and subsequently uses these features to predict task-agnostic sub-goals by reasoning about their latent relations. We evaluated ZAS-F in radio frequency identification (RFID) inventory tasks across various dynamic environments, a typical long-horizon task requiring robots to handle unpredictable conditions, including unseen objects and structural layouts. Our experiments demonstrated that ZAS-F achieves a learning efficiency 30 times higher than previous methods, requiring only 8k demonstrations. Compared to prior approaches, ZAS-F achieves a 98.3% scanning accuracy while significantly reducing the training data requirement. Further, ZAS-F demonstrated strong generalization, maintaining a scan success rate of 99.4% in real-world deployment without additional fine-tuning. In long-term operations spanning 100 rooms, ZAS-F maintained consistent performance compared to short-term tasks, highlighting its robustness against compounding e

关键词： Imitation Learning Long-horizon Task Reasoning RFID Inventory Sub-Goal Zero-shot Generalization

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：