检索结果-内蒙古大学图书馆

Transforming Experimental Cobot Cell to Industrial Realization – an Ethical AI Approach

IFAC-PapersOnLine 2023年第2期56卷 7335-7341页

作者： Istvan Mezgár József Váncza Imre Paniti József Tóth EPIC Center of Excellence in Production Informatics and Control Institute for Computer Science and Control Budapest Hungary Department of Manufacturing Science and Technology Budapest University of Technology and Economics Budapest Hungary Hepenix Ltd. Diósd Hungary

Collaborative Robot (cobot) cells are getting more and more integrative building blocks of Cyber Physical Enterprises. These cells integrate the advantages of human workers with the special capabilities of robots in a safe manner. Cobots need advanced, in many cases artificial intelligence (AI) based control systems to harmonize the collaborative activities. When transferring/transforming experimental setups into industrial application, not only technological and business related, but also ethical aspects have to be taken into consideration. The paper introduces a novel workflow supporting this transformation and presents its application in a case-study of a cobot cell which uses advanced sensing, symbolic AI planning and mixed reality techniques for planning and explaining visually the operation of the cell. The work which takes the responsible artificial intelligence (RAI) approach combines the actual relevant AI standards with the explicit requirements of industrial practitioners.

关键词： Collaborative robotics Artificial intelligence Ethics Design methodology

来源：评论

学校读者我要写书评

暂无评论

Novel heuristic approach to integrating task sequencing and production system configuration 55

Novel heuristic approach to integrating task sequencing and ...

引用

55th CIRP Conference on Manufacturing Systems, CIRP CMS 2022

作者： Tsutsumi, Daisuke Kovács, András Szalóki, Ádám Hitachi Ltd. Research and Development Group 292 Yoshida-chyo Totsuka-ku Yokohama-shi Kanagawa244-0817 Japan EPIC Centre of Excellence in Production Informatics and Control SZTAKI Kende 13-17 BudapestH-1111 Hungary

The robot system integration (SI) business targets the construction of production lines for various products that differ from customer to customer. Improving the efficiency of line design is important for increasing profits in the SI business. Especially in the construction of assembly lines, the optimal system configuration, with an appropriate combination of robotic and human stations, is critical for maximizing the return on investment of customers. Yet, the optimal system configuration depends heavily on the task sequence in the process plan. In this paper, task sequencing and system configuration are regarded as two interdependent sub-problems of line design. The exact joint optimum of the two problems can be found only if they are solved together, in an integrated way. However, this is computationally intractable for problems of industrially relevant sizes. To overcome this challenge, a heuristic method exploiting industrial knowledge is proposed for solving the two problems simultaneously. The heuristic constructs promising task sequences according to the capabilities of the applicable resources, considering aspects such as, among others, human work quality or robot tool change times. Then, a close-to-optimal system configuration is computed by considering the promising task sequences and their various relevant combinations. The proposed method is illustrated in a case study on the assembly of electrical components in the automotive industry. © 2022 The Authors. Published by Elsevier B.V.

关键词： assembly lines optimization system configuration task sequencing

来源：评论

学校读者我要写书评

暂无评论

Dynamic production Line Re-balancing by Alternative Plans for Compensating Equipment Failures 56

Dynamic Production Line Re-balancing by Alternative Plans fo...

引用

56th CIRP International Conference on Manufacturing Systems, CIRP CMS 2023

作者： Tsutsumi, Daisuke Bergmann, Júlia Dobrovoczki, Péter Hayashi, Naohiro Kovács, András Szádoczki, Zsombor Szaller, Ádám Umeda, Shota Hitachi Ltd. Research & Development Group 292 Yoshida-chyo Totsuka-ku Kanagawa Yokohama-shi244-0817 Japan EPIC Centre of Excellence in Production Informatics and Control SZTAKI Budapest Hungary Eötvös Loránd University Doctoral School of Informatics Budapest Hungary Eötvös Loránd University Doctoral School of Mathematics Budapest Hungary Department of Operations Research and Actuarial Sciences Corvinus University of Budapest Hungary EPIC InnoLabs Nonprofit Ltd. Budapest Hungary

As social uncertainty increases, there are changes in values such as safety, sustainability, and response to changes. The manufacturing industry is also required to improve its resilience by dynamically recombining production resources to respond to change. In the operation of a production line, if demand fluctuates due to changes in the external environment or customer needs, or if sudden production fluctuation such as equipment failure occurs, the production line will be stopped due to reconfiguration or equipment restoration. Therefore, the challenge is to continue production and improve productivity even when fluctuation occurs. In order to respond to the above challenges, this research aims to dynamically and quickly re-plan the production line using the existing equipment according to current equipment and production status. In this paper, failures of the production equipment, such as robots and tools, are regarded as production fluctuation factors. In order to continue production with the remaining resources in the event of equipment failure, it is necessary to change the process plan. However, since these are engineering tasks in the production preparation stage, there was the problem of not being able to flexibly change them at the manufacturing site during the production execution stage. In this development, an alternative plan pre-generation and selection method was developed to maintain production by re-allocating tasks during equipment failure. In this approach, first, multiple alternative plans for task allocation for failures in each piece of equipment are planned in advance using a process plan optimization technique. Next, production fluctuations due to equipment failures are detected. Finally, dynamic changes are possible by selecting alternative plans that maximize the throughput of the entire production system. Initial verification results are shown by comparison with the conventional method for equipment failure using simulation on a small-scale

关键词： alternative planning dynamic re-planning production line system configuration task allocation

来源：评论

学校读者我要写书评

暂无评论

Digital twin assisted workpiece referencing for compensating the stock deviation of casted parts

引用

Procedia CIRP 2023年 118卷 288-293页

作者： Tamás Cserteg Gábor Erdős Gergely Horváth András Kovács EPIC Centre of Excellence in Production Informatics and Control Institute for Computer Science and Control (SZTAKI) Budapest H-1111 Hungary Faculty of Informatics Eötvös Loránd University Budapest H-1117 Hungary Department of Manufacturing Science and Technology Budapest University of Technology and Economics Budapest H-1111 Hungary

Stock deviation of casted parts needs to be handled by adapting the CNC machining code to every new batch. Usually, human experts deal with this workpiece referencing task, but the demand for automation is expressed by the industry. This paper introduces a Digital Twin (DT) supported workpiece referencing method, implemented in the following steps: building the DT of the CNC machining cell, loading measurements of the casted part into the DT, solving the workpiece referencing problem as a convex optimization problem, and generating the compensated CNC code. The proposed approach is illustrated in a case study from the automotive industry.

关键词： Machining Workpiece referencing Digital Twin Convex optimization

来源：评论

学校读者我要写书评

暂无评论

Bilevel programming approach to demand response management with day-ahead tariff

引用

Journal of Modern Power Systems and Clean Energy 2019年第6期7卷 1632-1643页

作者： Andras KOVACS EPIC Center of Excellence in Production Informatics and Control Institute for Computer Science and ControlHungarian Academy of SciencesBudapestHungary

This paper introduces a bilevel programming approach to electricity tariff optimization for the purpose of demand response management(DRM)in smart *** the multi-follower Stackelberg game model,the leader is the profit-maximizing electricity retailer,who must set a time-of-use variable energy tariff in the *** correspond to the groups of prosumers(simultaneous producers and consumers of the electricity).They respond to the observed tariff,schedule controllable loads and determine the charging/discharging policy of their batteries to minimize the cost of electricity and to maximize the utility at the same time.A bilevel programming formulation of the problem is defined,and its fundamental properties are *** primal-dual reformulation is proposed in this paper to convert the bilevel optimization problem into a single-level quadratically constrained quadratic program(QCQP),and a successive linear programming(SLP)algorithm is applied to solve *** is demonstrated in computational experiments that the proposed approach outperforms earlier typical methods based on the KarushKuhn-Tucker(KKT)reformulation regarding both solution quality and computational efficiency on practically relevant problem ***,it also offers more flexible modeling capabilities.

关键词： Demand response management(DRM) Smart grid Game theory Optimization Bilevel programming

来源：评论

学校读者我要写书评

暂无评论

Tool failure recognition using inconsistent data

引用

Procedia CIRP 2022年 107卷 1204-1209页

作者： Júlia Bergmann Klaudia Éva Zeleny József Váncza Andrea Kő EPIC Center of Excellence in Production Informatics and Control Institute for Computer Science and Control (SZTAKI) Eötvös Loränd Research Network (ELKH) Budapest Hungary Eötvös Loränd University Doctoral School of Informatics Budapest Hungary EPIC InnoLabs Nonprofit Ltd. Budapest Hungary Corvinus University of Budapest Budapest Hungary Budapest University of Technology and Economics Dept. of Manufacturing Science and Technology Budapest Hungary

Data is everything - at least this is one of the main messages of the ongoing industrial revolution. Manufacturing companies all over the world are expanding their digital infrastructure and knowledge on data analysis in the hope of increasing their KPIs with the help of artificial intelligence (AI). Although several well-designed data-driven solutions are available, the most crucial part, data preparation is still not fully supported. In this paper a framework is presented for processing sensor data of machining processes with variable cycle times in an unstable environment. Traditional and novel AI algorithms are tested on the data of a vulcanization process from the automotive industry, namely from tire manufacturing’s curing phase. The process in question consists of several subprocesses, and the quality of curing is mostly dependent of the status of a specific type of machine tool. Conventional methods (e.g., examining the cured product manually) are currently used for failure recognition, however the examination is only feasible after a long delay due to the extreme level of heat, which leads to unnecessary and unwanted scrap production. Therefore, a more sophisticated and complex approach is required to increase quality score. A combination of mathematical methods is proposed combining t-SNE feature representation, convolutional neural network, and linear programming optimization. The model highly relies on the tool’s continuous degradation characteristics. The threshold for the given binary classification is set by maximizing the accuracy of the detection model. The main contribution of the research is the method of inconsistent sensor data manipulation which supports a unique combination of AI models for early failure recognition.

关键词： Type your keywords here separated by semicolons artificial intelligence failure detection data preparation t-SNE deep insight linear programming

来源：评论

学校读者我要写书评

暂无评论

Novel heuristic approach to integrating task sequencing and production system configuration

引用

Procedia CIRP 2022年 107卷 28-33页

作者： Daisuke Tsutsumi András Kovács Ádám Szalóki Hitachi Ltd. Research & Development Group 292 Yoshida-chyo Totsuka-ku Yokohama-shi Kanagawa 244-0817 Japan EPIC Centre of Excellence in Production Informatics and Control SZTAKI Kende 13-17 H-1111 Budapest Hungary

关键词： task sequencing system configuration assembly lines optimization

来源：评论

学校读者我要写书评

暂无评论

Dynamic production Line Re-balancing by Alternative Plans for Compensating Equipment Failures

引用

Procedia CIRP 2023年 120卷 810-815页

作者： Daisuke Tsutsumi Júlia Bergmann Péter Dobrovoczki Naohiro Hayashi András Kovács Zsombor Szádoczki Ádám Szaller Shota Umeda Hitachi Ltd. Research & Development Group 292 Yoshida-chyo Totsuka-ku Yokohama-shi Kanagawa 244-0817 Japan EPIC Centre of Excellence in Production Informatics and Control SZTAKI Budapest Hungary Eötvös Loránd University Doctoral School of Informatics Budapest Hungary EPIC InnoLabs Nonprofit Ltd. Budapest Hungary Eötvös Loránd University Doctoral School of Mathematics Budapest Hungary Department of Operations Research and Actuarial Sciences Corvinus University of Budapest Hungary

关键词： production line system configuration task allocation alternative planning dynamic re-planning

来源：评论

学校读者我要写书评

暂无评论

Sim2Real Grasp Pose Estimation for Adaptive Robotic Applications

引用

IFAC-PapersOnLine 2023年第2期56卷 5233-5239页

作者： Dániel Horváth Kristóf Bocsi Gábor Erdős Zoltán Istenes Centre of Excellence in Production Informatics and Control Institute for Computer Science and Control Eőtvős Loránd Research Network Budapest Hungary CoLocation Center for Academic and Industrial Cooperation Eőtvős Loránd University Budapest Hungary Department of Manufacturing Science and Engineering Budapest University of Technology and Economics Budapest Hungary

Adaptive robotics plays an essential role in achieving truly co-creative cyber physical systems. In robotic manipulation tasks, one of the biggest challenges is to estimate the pose of given workpieces. Even though the recent deep-learning-based models show promising results, they require an immense dataset for training. In this paper, two vision-based, multi-object grasp pose estimation models (MOGPE), the MOGPE Real-Time and the MOGPE High-Precision are proposed. Furthermore, a sim2real method based on domain randomization to diminish the reality gap and overcome the data shortage. Our methods yielded an 80% and a 96.67% success rate in a real-world robotic pick-and-place experiment, with the MOGPE Real-Time and the MOGPE High-Precision model respectively. Our framework provides an industrial tool for fast data generation and model training and requires minimal domain-specific data.

关键词： adaptive robotics robot vision sim2real knowledge transfer smart manufacturing cyber physical production systems

来源：评论

学校读者我要写书评

暂无评论

HiER: Highlight Experience Replay for Boosting Off-Policy Reinforcement Learning Agents

arXiv

引用

arXiv 2023年

作者： Horváth, Dániel Martín, Jesús Bujalance Erdos, Ferenc Gábor Istenes, Zoltán Moutarde, Fabien Center for Robotics MINES Paris PSL University Paris75272 France Centre of Excellence in Production Informatics and Control Institute for Computer Science and Control Hungarian Research Network Budapest1111 Hungary CoLocation Center for Academic and Industrial Cooperation Eötvös Loránd University Budapest1117 Hungary

Even though reinforcement-learning-based algorithms achieved superhuman performance in many domains, the field of robotics poses significant challenges as the state and action spaces are continuous, and the reward function is predominantly sparse. Furthermore, on many occasions, the agent is devoid of access to any form of demonstration. Inspired by human learning, in this work, we propose a method named highlight experience replay (HiER) that creates a secondary highlight replay buffer for the most relevant experiences. For the weights update, the transitions are sampled from both the standard and the highlight experience replay buffer. It can be applied with or without the techniques of hindsight experience replay (HER) and prioritized experience replay (PER). Our method significantly improves the performance of the state-of-the-art, validated on 8 tasks of three robotic benchmarks. Furthermore, to exploit the full potential of HiER, we propose HiER+ in which HiER is enhanced with an arbitrary data collection curriculum learning method. Our implementation, the qualitative results, and a video presentation are available on the project site: http://***/hier/. © 2023, CC BY-NC-ND.

关键词： Reinforcement learning

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：