检索结果-内蒙古大学图书馆

Grasping Force Estimation for Markerless Visuotactile Sensors

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Castaño-Amoros, Julio Gil, Pablo The AUROVA Lab Computer Science Research Institute University of Alicante Alicante03690 Spain The Department of Physics Systems Engineering and Signal Theory University of Alicante Alicante03690 Spain

Tactile sensors have been used for force estimation in the past, especially Vision-Based Tactile Sensors (VBTS) have recently become a new trend due to their high spatial resolution and low cost. In this work, we have designed and implemented several approaches to estimate the normal grasping force using different types of markerless visuotactile representations obtained from VBTS. Our main goal is to determine the most appropriate visuotactile representation, based on a performance analysis during robotic grasping tasks. Our proposal has been tested on the dataset generated with our DIGIT sensors and another one obtained using GelSight Mini sensors from another state-of-the-art work. We have also tested the generalization capabilities of our best approach, called RGBmod. The results led to two main conclusions. First, the RGB visuotactile representation is a better input option than the depth image or a combination of the two for estimating normal grasping forces. Second, RGBmod achieved a good performance when tested on 10 unseen everyday objects in real-world scenarios, achieving an average relative error of 0.125 ± 0.153. Furthermore, we show that our proposal outperforms other works in the literature that use RGB and depth information for the same task. © 2024, CC BY-NC-ND.

关键词： Depth perception

Integrating I4.0 Knowledge Graphs with Large Language Models Beyond SPARQL Endpoints 1

学校读者我要写书评

暂无评论

Integrating I4.0 Knowledge Graphs with Large Language Models...

Joint of the 1st Software Lifecycle Management for Knowledge Graphs Workshop and the 3rd International Workshop on Semantic Industrial Information Modelling, SOFLIM2KG-SEMIIM 2024

作者： Wahid, Abdul Yahya, Muhammad Zaman, Farooq Zhou, Baifan Breslin, John G. Intizar, Muhammad Ali Kharlamov, Evgeny Data Science Institute University of Galway IDA Business Park Lower Dangan GalwayH91 AEX4 Ireland Valeo Vision Systems Tuam Galways Ireland Information Technology University Lahore Pakistan Department of Computer Science Oslo Metropolitan University Oslo Norway School of Electronic Engineering Dublin City University Dublin Ireland Bosch Center for AI Renningen Germany

Industry 4.0 (I4.0) knowledge graphs are a common way to represent industrial information models. Conventional SPARQL querying systems require the users to be familiar with the data schema and SPARQL syntax. However, this is often very difficult for many users in industrial production, who have mostly an engineering background, instead of a semantic web. Recent developments in large language models (LLMs) make it possible for non-semantic experts to use natural language to query knowledge graphs (KG). In this work, we present a framework and preliminary results of integrating Industry 4.0 KGs with LLMs to improve how data is represented, reasoned, and processed in manufacturing contexts, facilitating user interaction with KGs and contributing to operational efficiency. Our technique enhances Language Models (LLMs) by utilising the semantic complexity and interdependence of Knowledge Graphs (KGs). This allows us to incorporate domain-specific knowledge. We used the FAISS library and LLaMA2 to optimise the storage and retrieval of vectors, which improved the system’s performance and scalability. This integration allows for advanced fault detection, proactive maintenance, and process optimisation, resulting in decreased periods of inactivity and improved productivity. We introduce the framework’s architecture, implementation strategy, and possible advantages while also discussing the difficulties associated with data integration and scalability. The results of our study show that the integration of KG-LLM surpasses traditional approaches in terms of operational efficiency, as evidenced by enhanced fault detection, proactive maintenance, and process optimisation, thereby opening up possibilities for the advancement of more intelligent and resilient production systems. © 2022 Copyright for this paper by its authors.

关键词： Knowledge graph

On the Effects and Optimal Design of Redundant Sensors in Collaborative State Estimation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Ren, Yunxiao Duan, Zhisheng Duan, Peihu Shi, Ling The State Key Laboratory for Turbulence and Complex Systems Department of Mechanics and Engineering Science College of Engineering Peking University Beijing100871 China School of Electrical Engineering and Computer Science KTH Royal Institute of Technology Stockholm Sweden The Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology Clear Water Bay Kowloon Hong Kong

The existence of redundant sensors in collaborative state estimation is a common occurrence, yet their true significance remains elusive. This paper comprehensively investigates the effects and optimal design of redundant sensors in sensor networks that use Kalman filtering to estimate the state of a random process collaboratively. The paper presents two main results: a theoretical analysis of the effects of redundant sensors and an engineering-oriented optimal design of redundant sensors. In the theoretical analysis, the paper leverages Riccati equations and Symplectic matrix theory to unveil the explicit role of redundant sensors in cooperative state estimation. The results unequivocally demonstrate that the addition of redundant sensors enhances the estimation performance of the sensor network, aligning with the principle of "more is better". Moreover, the paper establishes a precise sufficient and necessary condition to assess whether the inclusion of redundant sensors improves the overall estimation performance. Moving towards engineering-oriented design optimization, the paper proposes a novel algorithm to tackle the optimal design problem of redundant sensors, and the convergence of the proposed algorithm is guaranteed. Numerical simulations are provided to demonstrate the results. Copyright © 2023, The Authors. All rights reserved.

关键词： Kalman filters

On the Convergence of Gossip Learning in the Presence of Node Inaccessibility

学校读者我要写书评

暂无评论

On the Convergence of Gossip Learning in the Presence of Nod...

IEEE International Conference on Communications (ICC)

作者： Tian Liu Yue Cui Xueyang Hu Yecheng Xu Bo Liu Zhejiang Laboratory Hangzhou China Department of Computer Science and Computer Information Systems Auburn University at Montgomery Montgomery USA Department of Computer Science and Software Engineering Auburn University Auburn USA Institute of Rural Development Zhejiang Academy of Agricultural Sciences Hangzhou China DBAPPSecurity Co. Ltd. Hangzhou China

ISBN: (数字)9781728190549

ISBN: (纸本)9781728190556

Gossip learning (GL), as a decentralized alternative to federated learning (FL), is more suitable for resourceconstrained wireless networks, such as Flying Ad-Hoc Networks (FANETs) that are formed by unmanned aerial vehicles (UAVs). GL can significantly enhance the efficiency and extend the battery life of UAV networks. Despite the advantages, the performance of GL is strongly affected by data distribution, communication speed, and network connectivity. However, how these factors influence the GL convergence is still unclear. Existing work studied the convergence of GL based on a virtual quantity for the sake of convenience, which failed to reflect the real state of the network when some nodes are inaccessible. In this paper, we formulate and investigate the impact of inaccessible nodes to GL under a dynamic network topology. We first decompose the weight divergence by whether the node is accessible or not. Then, we investigate the GL convergence under the dynamic of node accessibility and theoretically provide how the number of inaccessible nodes, data non-i.i.d.-ness, and duration of inaccessibility affect the convergence. Extensive experiments are carried out in practical settings to comprehensively verify the correctness of our theoretical findings.

关键词： Training Accuracy Network topology Federated learning Wireless networks Autonomous aerial vehicles Ad hoc networks

Unearthing Undiscovered Interests: Knowledge Enhanced Representation Aggregation for Long-Tail Recommendation 1

学校读者我要写书评

暂无评论

10th International Symposium on Integrated Uncertainty in Knowledge Modelling and Decision Making, IUKM 2023

作者： Zhang, Zhipeng Zhang, Yuhang Hao, Tianyang Li, Zuoqing Zhang, Yao Inuiguchi, Masahiro School of Computer Science and Artificial Intelligence Liaoning Normal University Dalian China Dalian Gona Technology Group Co. Ltd. Dalian China School of Mechanical Engineering and Automation Dalian Polytechnic University Dalian China Department of Systems Innovation Osaka University Osaka Japan

ISBN: (数字)9783031467813

ISBN: (纸本)9783031467806

Graph neural networks have achieved remarkable performance in the field of recommender systems. However, existing graph-based recommendation approaches predominantly focus on suggesting popular items, disregarding the significance of long-tail recommendation and consequently falling short of meeting users’ personalized needs. To this end, we propose a novel approach called Knowledge-enhanced Representation Aggregation for Long-tail Recommendation (KRALR). Firstly, KRALR employs a user long-tail interests representation aggregation procedure to merge historical interaction information with rich semantic data extracted from knowledge graph (KG). By utilizing random walks on the KG and incorporating item popularity constraints, KRALR effectively captures the long-tail interests specific to the target user. Furthermore, KRALR introduces a long-tail item representation aggregation procedure by constructing a co-occurrence graph and integrating it with the KG. This integration enhances the quality of the representation for long-tail items, thereby enabling KRALR to provide more accurate recommendations. Finally, KRALR predicts rating scores for items that users have not interacted with and recommends the top N un-interacted items with the highest rating scores. Experimental results on the real-world dataset demonstrate that KRALR can improve recommendation accuracy and diversity simultaneously, and provide a wider array of satisfactory long-tail items for target users. Code is available at https://***/ZZP-RS/KRALR. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023.

关键词： Knowledge graph

GPU-Accelerated Privacy Aggregation Queries of Crowdsensing Spatio-temporal Data

学校读者我要写书评

暂无评论

IEEE Transactions on Consumer Electronics 2024年

作者： Li, Yuxi Chen, Jingjing Ji, Dong Zhu, Fa Chen, Xingchi Park, Jong Hyuk Northeastern University School of Computer Science and Engineering Shenyang110819 China Northeastern University National Frontiers Science Center for Industrial Intelligence and Systems Optimization Shenyang110819 China Nanjing Forestry University College of Information Sciences and Technology College of Artificial Intelligence Nanjing China Pengcheng Laboratory Shenzhen518000 China Department of Computer Science and Engineering 232 Gongneung-ro Seoul Korea Republic of

Addressing the challenges of query privacy leakage and high verification costs in multisource spatio-temporal data queries within crowdsensing environments, we present CrowdPQ-a novel privacy-preserving aggregation query scheme for spatio-temporal crowdsensing data utilizing GPU acceleration. Our approach leverages a robust two-server secure architecture. At the core of our scheme lies a privacy query protocol based on Function Secret Sharing (FSS), ensuring confidentiality while enabling efficient data aggregation queries. Additionally, we propose a non-interactive lightweight verification protocol utilizing Beaver multiplication triples, significantly minimizing resource consumption from invalid query tokens. We provide a comprehensive formal analysis of CrowdPQ in terms of correctness, security, and complexity. Our prototype, implemented in a two-server setting with GPU support, demonstrates superior performance: GPU acceleration reduces query processing time by 4×, server-side communication cost by 25×, and client-side communication by up to 88.89% compared with existing methods. These results highlight the potential of GPU-enabled secure processing for large-scale crowdsensing data aggregation tasks, offering an innovative pathway for privacy-focused data handling in complex environments. © 1975-2011 IEEE.

关键词： Spatio-temporal data

Risk-Averse Certification of Bayesian Neural Networks

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Xiyue Wang, Zifan Gao, Yulong Romao, Licio Abate, Alessandro Kwiatkowska, Marta Department of Computer Science University of Oxford United Kingdom Division of Decision and Control Systems KTH Royal Institute of Technology Sweden Department of Electrical and Electronic Engineering Imperial College London United Kingdom Department of Wind and Energy Systems Technical University of Denmark Denmark

In light of the inherently complex and dynamic nature of real-world environments, incorporating risk measures is crucial for the robustness evaluation of deep learning models. In this work, we propose a Risk-Averse Certification framework for Bayesian neural networks called RAC-BNN. Our method leverages sampling and optimisation to compute a sound approximation of the output set of a BNN, represented using a set of template polytopes. To enhance robustness evaluation, we integrate a coherent distortion risk measure—Conditional Value at Risk (CVaR)—into the certification framework, providing probabilistic guarantees based on empirical distributions obtained through sampling. We validate RAC-BNN on a range of regression and classification benchmarks and compare its performance with a state-of-the-art method. The results show that RAC-BNN effectively quantifies robustness under worst-performing risky scenarios, and achieves tighter certified bounds and higher efficiency in complex tasks. Copyright © 2024, The Authors. All rights reserved.

关键词： Risk assessment

Pixel Image Segmentation of Cerebral Hematoma Based on U-Net and Clustering

学校读者我要写书评

暂无评论

Pixel Image Segmentation of Cerebral Hematoma Based on U-Net...

computer, Vision and Intelligent Technology (ICCVIT), International Conference on

作者： Hai-Feng Zhu Yao-Hui Liu Kun Guo Si-Jie Zhang Ge Chen Department of Computer Science and Technology XiangNan University ChenZhou China Hunan Engineering Research Center of Advanced Embedded Computing and Intelligent Medical Systems Xiangnan University ChenZhou China

ISBN: (数字)9798331540043

ISBN: (纸本)9798331540050

Intracerebral hemorrhage is a disease with high fatality rate, and the expansion of hematoma caused by hemorrhage is one of the factors leading to poor prognosis. Fast and accurate segmentation of the contours of cerebral hematoma will help doctors to make diagnosis to a large extent. In order to solve the problem of insufficient precision of U-net segmentation, pixel image segmentation based on U-net and clustering is proposed. The CT images are divided into multiple channels by preprocessing, and on this basis, the cluster is processed twice, and then they enter the improved U-net network respectively. Finally, multiple results are fused to obtain the final segmentation result.

关键词： Image segmentation Accuracy Computed tomography Clustering algorithms Feature extraction Classification algorithms Kernel Hemorrhaging Unsupervised learning Diseases

Thermodynamic Interpolation: A generative approach to molecular thermodynamics and kinetics

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Moqvist, Selma Chen, Weilong Schreiner, Mathias Nüske, Feliks Olsson, Simon Department of Computer Science and Engineering Chalmers University of Technology University of Gothenburg GothenburgSE-41296 Sweden Max-Planck-Institute for Dynamics of Complex Technical Systems Magdeburg Germany

Using normalizing flows and reweighting, Boltzmann Generators enable equilibrium sampling from a Boltzmann distribution, defined by an energy function and thermodynamic state. In this work, we introduce Thermodynamic Interpolation (TI), which allows for generating sampling statistics in a temperature-controllable way. We introduce TI flavors that work directly in the ambient configurational space, mapping between different thermodynamic states or through a latent, normally distributed reference state. Our ambient-space approach allows for the specification of arbitrary target temperatures, ensuring generalizability within the temperature range of the training set and demonstrating the potential for extrapolation beyond it. We validate the effectiveness of TI on model systems that exhibit metastability and non-trivial temperature dependencies. Finally, we demonstrate how to combine TI-based sampling to estimate free energy differences through various free energy perturbation methods and provide corresponding approximated kinetic rates estimated through generator extended dynamic mode decomposition (gEDMD). Copyright © 2024, The Authors. All rights reserved.

关键词： Free energy