检索结果-内蒙古大学图书馆

model-based learning on state-based potential games for distributed self-optimization of manufacturing systems

JOURNAL OF MANUFACTURING SYSTEMS 2023年 71卷 474-493页

作者： Yuwono, Steve Schwung, Andreas South Westphalia Univ Appl Sci Dept Automat Technol & Learning Syst Ubecker Ring 2 D-59494 Soest Germany

In this paper, we propose a novel approach of model-based learning on state-based potential games (MB-SbPGs) that enables distributed self-optimization of manufacturing systems with a sample-efficient approach, where the interactions between the SbPG players and the actual systems can be significantly reduced. Most self-learning methods necessitate such interactions frequently to find the optimal solutions due to the exploratory and iterative behaviours of the players throughout the learning process. Nevertheless, in industrial environments, these interactions are generally expensive and time-consuming, as well as have potential risks. One possible approach to address this challenge is by employing an accurate digital representation of the system. However, developing an accurate digital representation is often a complex task. Therefore, the proposed MB-SbPG approach offers a solution that effectively learns the dynamics of the relevant systems and optimizes them within virtual environments. To achieve the stated objectives, the MB-SbPG approach consists of two primary steps. Firstly, deep learning models are designed to accurately predict system dynamics. Secondly, the SbPG players are trained within virtual environments, which enables the distributed self-optimization features of the system. Within the scope of this research, we examine two separate strategies for MB-SbPGs, namely (a) MB-SbPGs with single-step predictors and (b) MB-SbPGs with multi-step predictors. Furthermore, we demonstrate that MB-SbPGs have the potential to enhance memory-based learning on SbPG. We also validate the suitability of reusing the trained networks on MB-SbPGs for transfer learning in adaptable systems. To assess its effectiveness, we implement and validate MB-SbPG in a real industrial control scenario using a laboratory-scale testbed, which ends up with encouraging results.

关键词： model-based learning Game theory Machine learning Smart manufacturing Multi-step prediction Transfer learning

来源：评论

学校读者我要写书评

暂无评论

NeuPAN: Direct Point Robot Navigation With End-to-End model-based learning

引用

IEEE TRANSACTIONS ON ROBOTICS 2025年 41卷 2804-2824页

作者： Han, Ruihua Wang, Shuai Wang, Shuaijun Zhang, Zeqing Chen, Jianjun Lin, Shijie Li, Chengyang Xu, Chengzhong Eldar, Yonina C. Hao, Qi Pan, Jia Univ Hong Kong Dept Comp Sci Hong Kong 000000 Peoples R China Southern Univ Sci & Technol Dept Comp Sci & Engn Shenzhen 518055 Peoples R China Chinese Acad Sci Shenzhen Inst Adv Technol Shenzhen 100045 Peoples R China Hong Kong Univ Sci & Technol Guangzhou Robot & Autonomous Syst Thrust Guangzhou 511455 Peoples R China Univ Macau IOTSC Macau 999078 Peoples R China Weizmann Inst Sci IL-7621613 Rehovot Israel Southern Univ Sci & Technol Sifakis Res Inst Trustworthy Autonomous Syst Shenzhen 518055 Peoples R China

Navigating a nonholonomic robot in a cluttered, unknown environment requires accurate perception and precise motion control for real-time collision avoidance. This article presents neural proximal alternating-minimization network (NeuPAN): a real-time, highly accurate, map-free, easy-to-deploy, and environment-invariant robot motion planner. Leveraging a tightly coupled perception-to-control framework, NeuPAN has two key innovations compared to existing approaches: first, it directly maps raw point cloud data to a latent distance feature space for collision-free motion generation, avoiding error propagation from the perception to control pipeline;second, it is interpretable from an end-to-end model-based learning perspective. The crux of NeuPAN is solving an end-to-end mathematical model with numerous point-level constraints using a plug-and-play proximal alternating-minimization network, incorporating neurons in the loop. This allows NeuPAN to generate real-time, physically interpretable motions. It seamlessly integrates data and knowledge engines, and its network parameters can be fine-tuned via backpropagation. We evaluate NeuPAN on a ground mobile robot, a wheel-legged robot, and an autonomous vehicle, in extensive simulated and real-world environments. Results demonstrate that NeuPAN outperforms existing baselines in terms of accuracy, efficiency, robustness, and generalization capabilities across various environments, including the cluttered sandbox, office, corridor, and parking lot. We show that NeuPAN works well in unknown and unstructured environments with arbitrarily shaped objects, transforming impassable paths into passable ones.

关键词： Robots Navigation Collision avoidance Mobile robots Mathematical models Accuracy Autonomous vehicles Approximation algorithms Robot sensing systems Real-time systems Direct point robot navigation model-based learning optimization-based collision avoidance (OBCA)

来源：评论

学校读者我要写书评

暂无评论

The neural substrates of how model-based learning affects risk taking: Functional coupling between right cerebellum and left caudate

引用

BRAIN AND COGNITION 2023年 172卷 106088页

作者： Huo, Hangfeng Lesage, Elise Dong, Wenshan Verguts, Tom Seger, Carol A. Diao, Sitong Feng, Tingyong Chen, Qi Guangxi Normal Univ Fac Educ Dept Psychol Guilin Peoples R China Shenzhen Univ Sch Psychol Shenzhen 518060 Peoples R China South China Normal Univ Sch Psychol Guangzhou 510631 Peoples R China South China Normal Univ Ctr Studies Psychol Applicat Guangzhou 510631 Peoples R China South China Normal Univ Guangdong Key Lab Mental Hlth & Cognit Sci Guangzhou 510631 Peoples R China Univ Ghent Dept Expt Psychol Ghent Belgium Southwest Univ Fac Psychol Res Ctr Psychol & Social Dev Chongqing Peoples R China Minist Educ Key Lab Cognit & Personal Chongqing Peoples R China Colorado State Univ Dept Psychol Ft Collins CO 80523 USA Colorado State Univ Program Mol Cellular & Integrat Neurosci Ft Collins CO 80523 USA Shenzhen Univ Sch Psychol Shenzhen Peoples R China

Higher executive control capacity allows people to appropriately evaluate risk and avoid both excessive risk aversion and excessive risk-taking. The neural mechanisms underlying this relationship between executive function and risk taking are still unknown. We used voxel-based morphometry (VBM) analysis combined with resting-state functional connectivity (rs-FC) to evaluate how one component of executive function, model-based learning, relates to risk taking. We measured individuals' use of the model-based learning system with the twostep task, and risk taking with the Balloon Analogue Risk Task. Behavioral results indicated that risk taking was positively correlated with the model-based weighting parameter omega. The VBM results showed a positive association between model-based learning and gray matter volume in the right cerebellum (RCere) and left inferior parietal lobule (LIPL). Functional connectivity results suggested that the coupling between RCere and the left caudate (LCAU) was correlated with both model-based learning and risk taking. Mediation analysis indicated that RCere-LCAU functional connectivity completely mediated the effect of model-based learning on risk taking. These results indicate that learners who favor model-based strategies also engage in more appropriate risky behaviors through interactions between reward-based learning, error-based learning and executive control subserved by a caudate, cerebellar and parietal network.

关键词： Decision making Risk taking model-based learning Functional connectivity Resting-state fMRI

来源：评论

学校读者我要写书评

暂无评论

Predicting subgrain size and dislocation density in machining-induced surface microstructure of nickel using supervised model-based learning

引用

MATERIALS TODAY COMMUNICATIONS 2022年 30卷

作者： Faruk Abedrabbo, Anibal Camilo Osorio, Juan Abolghasem, Sepideh Valencia, Carlos Rojas, Fabio Univ Los Andes Dept Ind Engn Carrera 1a 18A-10 Bogota 111711 Colombia Univ Los Andes Dept Mech Engn Carrera 1a 18A-10 Bogota 111711 Colombia

Microstructure transformations during severe shear deformation induced by large strain machining experiments are examined on high-purity (99.99%) nickel plastically deformed to strain (1-5) and strain-rate (10(2)-10(4)/s). Deformation conditions are created using plane strain machining (PSM). They are characterized via in-situ techniques, which are then juxtaposed with orientation imaging microscopy (OIM) via electron backscattered diffraction (EBSD), and the dislocation densities are quantified using the broadening of X-ray diffraction peaks of crystallographic planes. We parameterize the variation of microstructure response by measuring the subgrain size as a function of seven variables involved in the cutting process: cutting speed, rake angle, temperature, strain, strain-rate, Zenner-Holloman parameter, and a dependent rate parameter (R). This parametrization was based on supervised model-based learning. One of them used Principal Component Analysis (PCA) within a linear regression, which produces good predictions. We use the PCA model in addition to a FEM simulation of the PSM to predict the subsurface subgrain size. Furthermore, the Principle of Similitude (PS) is incorporated for predicting the dislocation densities in both the deformed chip and the machined subsurface. The proposed framework (FEM simulation-PCA model-PS relation) is shown to offer opportunities for creating multifunctional surface microstructure in an array of machining manufacturing processes.

关键词： model-based learning Nickel Subgrain size Dislocation density Orthogonal cutting surface SPD severe plastic deformation

来源：评论

学校读者我要写书评

暂无评论

Efficient Frequency Selective Surface Analysis via End-to-End model-based learning 19

Efficient Frequency Selective Surface Analysis via End-to-En...

引用

19th European Conference on Antennas and Propagation, EuCAP 2025

作者： Hammami, Cheima Polo-López, Lucas Magoarou, Luc Le Univ Rennes Insa Rennes Cnrs IETR-UMR 6164 Rennes France

ISBN: (纸本)9788831299107

This paper introduces an innovative end-to-end model-based deep learning approach for efficient electromagnetic analysis of high-dimensional frequency selective surfaces (FSS). Unlike traditional data-driven methods that require large datasets, this approach combines physical insights from equivalent circuit models with deep learning techniques to significantly reduce model complexity and enhance prediction accuracy. Compared to previously introduced model-based learning approaches, the proposed method is trained end - to-end from the physical structure of the FSS (geometric parameters) to its electromagnetic response (S-parameters). Additionally, an improvement in phase prediction accuracy through a modified loss function is presented. Comparisons with direct models, including deep neural networks (DNN) and radial basis function networks (RBFN), demonstrate the superiority of the model-based approach in terms of computational efficiency, model size, and generalization capability. © 2025 European Association on Antennas and Propagation.

关键词： Deep learning Frequency Selective Surface (FSS) model-based learning S-parameters

来源：评论

学校读者我要写书评

暂无评论

VeriSIM: A model-based learning pedagogy for fostering software design evaluation skills in computer science undergraduates

引用

RESEARCH AND PRACTICE IN TECHNOLOGY ENHANCED learning 2022年第1期17卷 1-35页

作者： Prasad, Prajish Iyer, Sridhar Indian Inst Technol Interdisciplinary Programme Educ Technol Mumbai Maharashtra India

Evaluating a software design is an important practice of expert software designers. They spend significant time evaluating their solution, by developing an integrated mental model of the software design and the requirements. However, sufficient emphasis has not been given on teaching and learning of evaluation practices in software design courses, and hence, graduating students find it difficult to critically analyse an existing design and improve upon it. In this paper, we describe a model-based learning pedagogy for teaching-learning of software design evaluation. model-based learning has been extensively used in science education and entails helping students construct, refine, revise, evaluate, and validate scientific models. We argue that modelling practices in software design evaluation are analogous to these practices. We adapted the model-based learning paradigm and operationalised it into a technology-enhanced learning environment (TELE) for fostering software design evaluation skills in computer science undergraduates. We conducted a research study with 22 undergraduate students to explore how the TELE and its features help students effectively evaluate a given software design. Students attempted a pre-test and post-test which asked them to identify defects in the design. We used the content analysis method to identify categories of defects from student responses in the pre-test and post-test. We also analysed student interaction logs and conducted focus group interviews to identify how features in the TELE contributed towards student learning. Findings from the study showed that students' understanding of evaluation improved, from merely adding new functionalities and requirements, to a process which involved identifying alternate scenarios in the design which violate the given requirements. Students perceived that pedagogical features of the TELE were useful in helping them effectively evaluate software designs. Findings from the study provide evidence fo

关键词： model-based learning UML diagrams Software design evaluation Scenario modelling and tracing Content analysis Thematic analysis

来源：评论

学校读者我要写书评

暂无评论

A Platform for model-based learning and Gamification in Design Education

引用

Procedia CIRP 2024年 128卷 7-12页

作者： Pooja Shenoy Tarun Kumar PES University Bangalore 560085 India Indian Institute of Science Bangalore 560012 India

This study proposes a design learning platform that combines concepts from gamification, model-based learning, and educational psychology to produce a dynamic learning environment created for design students. The platform’s main feature is a gamified approach to design education. It presents students with design quests and challenges that cross a variety of design disciplines. By using a subjective and open-ended approach, this platform encourages problem-solving, and critical thinking. To improve the learning experience, the platform incorporates educational psychology. Real-time feedback mechanisms are integrated with formative assessment methodologies to give students feedback on their progress. model-based learning tools, such as concept models, 3D design, and flowcharts are incorporated. This helps students in understanding the connections between ideas, visualising complicated design concepts, and applying their learning to real-world situations. Gamification components such as point systems and prizes help to increase motivation and engagement. The platform aims to combine Augmented reality (AR) and Virtual reality (VR) technology into the concept. Students will be able to explore design environments and develop interesting concepts. This advancement will increase the depth and interaction of the learning experience. Primary research was conducted with preliminary prototypes of the platform with students of design to recognise the initial impact of the features of the platform. This showed to improve design thinking and problem-solving abilities. Prototype concepts have been discussed to further develop the platform’s capabilities. These concepts include the development of a mobile application to facilitate skill tracking and daily design challenges to encourage innovation. In essence, this design learning platform provides a systemic approach to design education, combining various design methods and principles, educational psychology techniques, and AR and V

关键词： Gamification model-based learning Design education AR/VR

来源：评论

学校读者我要写书评

暂无评论

A model-based reinforcement learning framework for building heating management with branched rollout strategy and time-series prediction model

引用

BUILDING SIMULATION 2025年 1-20页

作者： Qu, Kaichen Zhang, Hong Zhou, Xin Ferrando, Martina Causone, Francesco Southeast Univ Sch Architecture Si Pai Lou 2 Nanjing 210096 Peoples R China Politecn Milan Dept Energy Via Lambruschini 4 I-20156 Milan Italy

Reinforcement learning (RL) has emerged as a promising approach for building energy management (BEM). However, most existing research focuses on model-free reinforcement learning (MFRL) approaches, which can encounter the learning challenge for heating, ventilation and air conditioning (HVAC) control due to extensive trial-and-error explorations and lengthy training times. To address this challenge, we propose a model-based reinforcement learning (MBRL) framework that incorporates a virtual environment to augment the agent's exploration. By leveraging the branched rollout strategy to generate short rollout predictions branched from the experience trajectory, the MBRL method mitigates compounding errors introduced by the time-series prediction model, enabling robust and efficient policy updates. Evaluated in an EnergyPlus testbed with real-world data verification, the proposed method demonstrates significant advantages: (1) RL-based controllers outperform the rule-based control (RBC) baseline after one training episode, (2) MBRL reduces training time by over 50% compared to MFRL while maintaining comparable control performance, and (3) an equal mix of real and synthetic data for MBRL training achieves an optimal trade-off between efficiency and control outcomes. This study contributes an efficient model-based training method for RL development in HVAC control, offering insights into advanced control strategies for BEM applications.

关键词： building energy management reinforcement learning model-based learning recursive prediction time-series prediction model

来源：评论

学校读者我要写书评

暂无评论

Integrating socioscientific issues and model-based learning to decide on a local issue: the white butterfly unit

引用

SCIENCE ACTIVITIES-PROJECTS AND CURRICULUM IDEAS IN STEM CLASSROOMS 2023年第2期60卷 90-105页

作者： Durak, Benzegul Topcu, Mustafa Sami Duzce Univ Dept Math & Sci Educ Duzce Turkiye Yildiz Tech Univ Dept Math & Sci Educ Istanbul Turkiye

Recent research suggests that integrating model-based learning and socioscientific issue based instruction helps students construct meaningful learning in science classrooms. Thus, this paper presents a unit plan that integrates model-based learning and socioscientific issues. The focus of the unit is the white butterfly which is a local pest. A 4-week middle school unit was developed on the driving question, "How might the population change of the white butterflies interact with the local ecosystems?". Evaluation of the unit was made through students' models and explanations. There were three modeling sessions throughout the unit. "Components" and "Explanatory Process" rubrics were developed to evaluate the models and explanations. The results showed that students' pre-knowledge about the white butterfly and its relationship with the local ecosystem was quite limited and throughout the lessons they were questioned and gained knowledge about the relationship between the white butterfly and ecosystem. Also, students became able to explain the effects and results of the methods used to control the white butterfly but not able to show them by drawing. Therefore, it is highly recommended to evaluate the models together with their explanations, as students' drawing skills are a limitation for the use of models.

关键词： socioscientific issues SSI-based instruction modeling model-based learning

来源：评论

学校读者我要写书评

暂无评论

model-based Online learning for Active ISAC Waveform Optimization

引用

IEEE JOURNAL OF SELECTED TOPICS IN SIGNAL PROCESSING 2024年第5期18卷 737-751页

作者： Pulkkinen, Petteri Koivunen, Visa Aalto Univ Dept Informat & Commun Engn FI-02150 Espoo Finland Saab Finland Oy FI-00100 Helsinki Finland

This paper proposes a model-based Online learning (MBOL) framework for waveform optimization in integrated sensing and communications (ISAC) systems. In particular, the MBOL framework is proposed to enhance the ISAC performance under dynamic environmental conditions. Unlike model-Free Online learning (MFOL) methods, our approach leverages a rich structural knowledge of sensing, communications, and radio environments, offering better explainability and sample efficiency. This paper establishes a theoretical analysis of the proposed class of MBOL methods, showing essential performance conditions and convergence rates. This theoretical analysis is critical for understanding the potential of MBOL in active waveform optimization tasks. We demonstrate the proposed MBOL framework in multicarrier ISAC systems, focusing on the sub-carrier selection and power allocation problem. Via numerical experiments, we show that the proposed MBOL method outperforms the MFOL method in terms of sample efficiency. The results underline the potential of MBOL for improving the active waveform optimization performance in ISAC systems, particularly when sample efficiency and explainability are critical.

关键词： Sensors Resource management Radar Task analysis Signal processing algorithms Markov decision processes Interference Joint radar-communications systems reinforcement learning online convex optimization model-based learning resource allocation waveform optimization

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：