Robotics and Computer-integrated Manufacturing最新文献_第7页

Automating the hand layup process: On the removal of protective films with collaborative robots 手铺过程的自动化：用协作机器人去除保护膜

IF 9.1 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Robotics and Computer-integrated Manufacturing

Pub Date : 2024-12-04 DOI: 10.1016/j.rcim.2024.102899

Renat Kermenov , Sergi Foix , Júlia Borràs , Vincenzo Castorani , Sauro Longhi , Andrea Bonci

This paper explores the issue of protective film removal in the hand layup process for composite parts production. The hand layup process, involving the assembly of prepreg plies onto a mold, is a skill-intensive task performed by multiple expert workers. A significant limitation of this method is its low repeatability, which impacts both the consistency and quality of the final product. The current research trend has the objective of developing autonomous or semi-autonomous layup cells to enhance process consistency, reduce production costs, and improve product quality.

Despite all this interest in bringing automation in composite manufacturing, an area left relatively unexplored is the removal of protective films from prepregs. The plies used in the hand layup process, are generally covered by those films that are removed by the workers during the manual layup activity. The manual removal of protective films from prepregs is a tedious and valueless task, which represents a bottleneck in achieving full or semi-automation of the layup process. For this reason, an autonomous or semi-autonomous cell needs to perform it to be market-relevant.

In this work, we propose a new effective method for initiating the peeling and integrate this method into a complete framework for the removal of protective films. This solution is designed to be easily integrated into a variety of existing cells. Finally, we validate our framework with an experimental proof of concept (PoC) which makes use of two collaborative robots for task execution.

本文探讨了复合材料零件手工铺层过程中保护膜的去除问题。手工铺层过程，包括将预浸料层组装到模具上，是一项由多名专业工人完成的技能密集型任务。该方法的一个重要局限性是重复性低，这影响了最终产品的一致性和质量。目前的研究趋势是开发自主或半自主的层叠单元，以提高工艺一致性，降低生产成本，提高产品质量。

{"title":"Automating the hand layup process: On the removal of protective films with collaborative robots","authors":"Renat Kermenov , Sergi Foix , Júlia Borràs , Vincenzo Castorani , Sauro Longhi , Andrea Bonci","doi":"10.1016/j.rcim.2024.102899","DOIUrl":"10.1016/j.rcim.2024.102899","url":null,"abstract":"<div><div>This paper explores the issue of protective film removal in the hand layup process for composite parts production. The hand layup process, involving the assembly of prepreg plies onto a mold, is a skill-intensive task performed by multiple expert workers. A significant limitation of this method is its low repeatability, which impacts both the consistency and quality of the final product. The current research trend has the objective of developing autonomous or semi-autonomous layup cells to enhance process consistency, reduce production costs, and improve product quality.</div><div>Despite all this interest in bringing automation in composite manufacturing, an area left relatively unexplored is the removal of protective films from prepregs. The plies used in the hand layup process, are generally covered by those films that are removed by the workers during the manual layup activity. The manual removal of protective films from prepregs is a tedious and valueless task, which represents a bottleneck in achieving full or semi-automation of the layup process. For this reason, an autonomous or semi-autonomous cell needs to perform it to be market-relevant.</div><div>In this work, we propose a new effective method for initiating the peeling and integrate this method into a complete framework for the removal of protective films. This solution is designed to be easily integrated into a variety of existing cells. Finally, we validate our framework with an experimental proof of concept (PoC) which makes use of two collaborative robots for task execution.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"93 ","pages":"Article 102899"},"PeriodicalIF":9.1,"publicationDate":"2024-12-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142788901","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A deep learning-enabled visual-inertial fusion method for human pose estimation in occluded human-robot collaborative assembly scenarios 基于深度学习的人机协同装配场景中人体姿态估计的视觉惯性融合方法

IF 9.1 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Robotics and Computer-integrated Manufacturing

Pub Date : 2024-11-30 DOI: 10.1016/j.rcim.2024.102906

Baicun Wang , Ci Song , Xingyu Li , Huiying Zhou , Huayong Yang , Lihui Wang

In the context of human-centric smart manufacturing, human-robot collaboration (HRC) systems leverage the strengths of both humans and machines to achieve more flexible and efficient manufacturing. In particular, estimating and monitoring human motion status determines when and how the robots cooperate. However, the presence of occlusion in industrial settings seriously affects the performance of human pose estimation (HPE). Using more sensors can alleviate the occlusion issue, but it may cause additional computational costs and lower workers' comfort. To address this issue, this work proposes a visual-inertial fusion-based method for HPE in HRC, aiming to achieve accurate and robust estimation while minimizing the influence on human motion. A part-specific cross-modal fusion mechanism is designed to integrate spatial information provided by a monocular camera and six Inertial Measurement Units (IMUs). A multi-scale temporal module is developed to model the motion dependence between frames at different granularities. Our approach achieves 34.9 mm Mean Per Joint Positional Error (MPJPE) on the TotalCapture dataset and 53.9 mm on the 3DPW dataset, outperforming state-of-the-art visual-inertial fusion-based methods. Tests on a synthetic-occlusion dataset further validate the occlusion robustness of our network. Quantitative and qualitative experiments on a real assembly case verified the superiority and potential of our approach in HRC. It is expected that this work can be a reference for human motion perception in occluded HRC scenarios.

在以人为中心的智能制造背景下，人机协作（HRC）系统利用人和机器的优势来实现更灵活和高效的制造。特别是，对人类运动状态的估计和监控决定了机器人何时以及如何合作。然而，工业环境中遮挡的存在严重影响了人体姿态估计（HPE）的性能。使用更多的传感器可以缓解遮挡问题，但它可能会导致额外的计算成本和降低工人的舒适度。为了解决这一问题，本研究提出了一种基于视觉-惯性融合的HRC HPE方法，旨在实现准确和鲁棒的估计，同时最大限度地减少对人体运动的影响。设计了一个部件特定的跨模态融合机制，以整合由单目相机和六个惯性测量单元（imu）提供的空间信息。开发了一个多尺度时间模块来模拟不同粒度帧之间的运动依赖关系。我们的方法在TotalCapture数据集上实现了34.9 mm的平均每个关节位置误差（MPJPE），在3DPW数据集上实现了53.9 mm，优于最先进的基于视觉惯性融合的方法。在合成遮挡数据集上的测试进一步验证了我们网络的遮挡鲁棒性。一个实际装配案例的定量和定性实验验证了该方法在HRC中的优越性和潜力。期望这项工作可以为人类在闭塞HRC场景下的运动感知提供参考。

{"title":"A deep learning-enabled visual-inertial fusion method for human pose estimation in occluded human-robot collaborative assembly scenarios","authors":"Baicun Wang , Ci Song , Xingyu Li , Huiying Zhou , Huayong Yang , Lihui Wang","doi":"10.1016/j.rcim.2024.102906","DOIUrl":"10.1016/j.rcim.2024.102906","url":null,"abstract":"<div><div>In the context of human-centric smart manufacturing, human-robot collaboration (HRC) systems leverage the strengths of both humans and machines to achieve more flexible and efficient manufacturing. In particular, estimating and monitoring human motion status determines when and how the robots cooperate. However, the presence of occlusion in industrial settings seriously affects the performance of human pose estimation (HPE). Using more sensors can alleviate the occlusion issue, but it may cause additional computational costs and lower workers' comfort. To address this issue, this work proposes a visual-inertial fusion-based method for HPE in HRC, aiming to achieve accurate and robust estimation while minimizing the influence on human motion. A part-specific cross-modal fusion mechanism is designed to integrate spatial information provided by a monocular camera and six Inertial Measurement Units (IMUs). A multi-scale temporal module is developed to model the motion dependence between frames at different granularities. Our approach achieves 34.9 mm Mean Per Joint Positional Error (MPJPE) on the TotalCapture dataset and 53.9 mm on the 3DPW dataset, outperforming state-of-the-art visual-inertial fusion-based methods. Tests on a synthetic-occlusion dataset further validate the occlusion robustness of our network. Quantitative and qualitative experiments on a real assembly case verified the superiority and potential of our approach in HRC. It is expected that this work can be a reference for human motion perception in occluded HRC scenarios.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"93 ","pages":"Article 102906"},"PeriodicalIF":9.1,"publicationDate":"2024-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142747246","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A self-imitation learning approach for scheduling evaporation and encapsulation stages of OLED display manufacturing systems 有机发光二极管显示制造系统蒸发和封装阶段调度的自模仿学习方法

IF 9.1 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Robotics and Computer-integrated Manufacturing

Pub Date : 2024-11-29 DOI: 10.1016/j.rcim.2024.102917

Donghun Lee , In-Beom Park , Kwanho Kim

In modern organic light-emitting diode (OLED) manufacturing systems, scheduling is a key decision-making problem to improve productivity. In particular, the scheduling of evaporation and encapsulation stages has been confronted with complicated constraints such as job-splitting property, preventive maintenance, machine eligibility, family setups, and heterogeneous release time of jobs. To efficiently solve such complicated scheduling problems, reinforcement learning (RL) has drawn increasing attention as an alternative in recent years. Unfortunately, the performance of the RL-based scheduling methods might not be satisfactory since unexpected correlations between actions are caused by machine eligibility restrictions, making it more challenging to address the credit assignment problem. To minimize the total tardiness, this article proposes a self-imitation learning-based scheduling method in which an agent utilizes past good experiences to exploit efficient exploration. Furthermore, a novel return design is introduced to overcome the credit assignment problem by considering machine eligibility restrictions. To prove the effectiveness and efficiency of the proposed method, numerical experiments are carried out by using the datasets that simulated the real-world OLED display manufacturing systems. Experiment results demonstrate that the proposed method outperforms other baselines, including rule-based and meta-heuristics, as well as the other DRL-based method in terms of the total tardiness while reducing computation time compared to meta-heuristics.

在现代有机发光二极管（OLED）制造系统中，调度是提高生产效率的关键决策问题。特别是，蒸发和封装阶段的调度面临着作业拆分性、预防性维护、机器合格性、家庭设置和作业异构释放时间等复杂的约束。为了有效地解决这些复杂的调度问题，强化学习（RL）作为一种替代方法近年来受到越来越多的关注。不幸的是，基于rl的调度方法的性能可能不令人满意，因为操作之间的意外关联是由机器资格限制引起的，这使得解决信用分配问题更具挑战性。为了最大限度地减少总延误，本文提出了一种基于自我模仿学习的调度方法，其中智能体利用过去的良好经验进行有效的探索。在此基础上，引入了一种新的回归设计，通过考虑机器的资格限制来克服信用分配问题。为了验证所提方法的有效性和高效性，利用模拟真实OLED显示制造系统的数据集进行了数值实验。实验结果表明，该方法在总延迟方面优于其他基准方法，包括基于规则和元启发式方法，以及其他基于drl的方法，同时与元启发式方法相比减少了计算时间。

{"title":"A self-imitation learning approach for scheduling evaporation and encapsulation stages of OLED display manufacturing systems","authors":"Donghun Lee , In-Beom Park , Kwanho Kim","doi":"10.1016/j.rcim.2024.102917","DOIUrl":"10.1016/j.rcim.2024.102917","url":null,"abstract":"<div><div>In modern organic light-emitting diode (OLED) manufacturing systems, scheduling is a key decision-making problem to improve productivity. In particular, the scheduling of evaporation and encapsulation stages has been confronted with complicated constraints such as job-splitting property, preventive maintenance, machine eligibility, family setups, and heterogeneous release time of jobs. To efficiently solve such complicated scheduling problems, reinforcement learning (RL) has drawn increasing attention as an alternative in recent years. Unfortunately, the performance of the RL-based scheduling methods might not be satisfactory since unexpected correlations between actions are caused by machine eligibility restrictions, making it more challenging to address the credit assignment problem. To minimize the total tardiness, this article proposes a self-imitation learning-based scheduling method in which an agent utilizes past good experiences to exploit efficient exploration. Furthermore, a novel return design is introduced to overcome the credit assignment problem by considering machine eligibility restrictions. To prove the effectiveness and efficiency of the proposed method, numerical experiments are carried out by using the datasets that simulated the real-world OLED display manufacturing systems. Experiment results demonstrate that the proposed method outperforms other baselines, including rule-based and meta-heuristics, as well as the other DRL-based method in terms of the total tardiness while reducing computation time compared to meta-heuristics.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"93 ","pages":"Article 102917"},"PeriodicalIF":9.1,"publicationDate":"2024-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142747248","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Digital twin-driven virtual commissioning for robotic machining enhanced by machine learning 机器学习增强的机器人加工数字双驱动虚拟调试

IF 9.1 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Robotics and Computer-integrated Manufacturing

Pub Date : 2024-11-29 DOI: 10.1016/j.rcim.2024.102908

Hepeng Ni , Tianliang Hu , Jindong Deng , Bo Chen , Shuangsheng Luo , Shuai Ji

Robotic machining has been increasingly applied in intelligent manufacturing production lines. Compared with the traditional machine tools, commissioning for robotic machining system (RMS) is particularly important due to the low accuracy of industrial robots (IRs). Traditional site commissioning has large workload and is difficult to handle the multi-source errors. Since digital twin (DT) provides strategies for staying synchronized with the physical entities in whole lifecycle, a DT-driven virtual commissioning (VC) system for RMS is developed in this study to improve machining accuracy and reduce the difficulty of commissioning. Firstly, the framework of DT-driven VC system is designed including several function modules such as interaction, data pre-processing, DT model of RMS (RMSDT), and optimization service. Since RMSDT is the kernel of precise VC, a machine learning-enhanced RMSDT oriented to actual machining path prediction is then constructed based on a proposed joint error equivalent strategy, which can fully consider the coupled multi-source errors of machining robot. After that, a practical consistency retention method for RMSDT is proposed based on a stepwise updating strategy, where the model performance can be maintained with low updating costs. Finally, a visual VC system is developed for the experimental 6-degree of freedom robotic milling platform to verify the feasibility and effectiveness of the VC framework. Multiple experiments are also performed to test the performance of RMSDT and contour error compensation. This study has useful reference for the enterprises engaged in RMS and has positive significance for promoting the robotic machining.

机器人加工在智能制造生产线上的应用越来越广泛。与传统机床相比，由于工业机器人的精度较低，机器人加工系统（RMS）的调试尤为重要。传统的现场调试工作量大，多源误差难以处理。由于数字孪生（DT）提供了在全生命周期内与物理实体保持同步的策略，为了提高加工精度和降低调试难度，本研究开发了一个数字孪生驱动的RMS虚拟调试（VC）系统。首先，设计了DT驱动VC系统的框架，包括交互、数据预处理、RMS的DT模型（RMSDT）和优化服务等功能模块；由于RMSDT是精密VC的核心，在提出的关节误差等效策略的基础上，构建了面向实际加工路径预测的机器学习增强RMSDT，充分考虑了加工机器人的耦合多源误差。在此基础上，提出了一种实用的基于逐步更新策略的RMSDT一致性保持方法，该方法可以以较低的更新成本保持模型的性能。最后，针对实验六自由度机器人铣削平台开发了可视化的VC系统，验证了VC框架的可行性和有效性。通过多次实验验证了RMSDT和轮廓误差补偿的性能。本研究对从事RMS的企业具有有益的借鉴意义，对推动机器人加工具有积极的意义。

{"title":"Digital twin-driven virtual commissioning for robotic machining enhanced by machine learning","authors":"Hepeng Ni , Tianliang Hu , Jindong Deng , Bo Chen , Shuangsheng Luo , Shuai Ji","doi":"10.1016/j.rcim.2024.102908","DOIUrl":"10.1016/j.rcim.2024.102908","url":null,"abstract":"<div><div>Robotic machining has been increasingly applied in intelligent manufacturing production lines. Compared with the traditional machine tools, commissioning for robotic machining system (RMS) is particularly important due to the low accuracy of industrial robots (IRs). Traditional site commissioning has large workload and is difficult to handle the multi-source errors. Since digital twin (DT) provides strategies for staying synchronized with the physical entities in whole lifecycle, a DT-driven virtual commissioning (VC) system for RMS is developed in this study to improve machining accuracy and reduce the difficulty of commissioning. Firstly, the framework of DT-driven VC system is designed including several function modules such as interaction, data pre-processing, DT model of RMS (RMSDT), and optimization service. Since RMSDT is the kernel of precise VC, a machine learning-enhanced RMSDT oriented to actual machining path prediction is then constructed based on a proposed joint error equivalent strategy, which can fully consider the coupled multi-source errors of machining robot. After that, a practical consistency retention method for RMSDT is proposed based on a stepwise updating strategy, where the model performance can be maintained with low updating costs. Finally, a visual VC system is developed for the experimental 6-degree of freedom robotic milling platform to verify the feasibility and effectiveness of the VC framework. Multiple experiments are also performed to test the performance of RMSDT and contour error compensation. This study has useful reference for the enterprises engaged in RMS and has positive significance for promoting the robotic machining.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"93 ","pages":"Article 102908"},"PeriodicalIF":9.1,"publicationDate":"2024-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142747247","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Efficient tool path planning method of ball-end milling for high quality manufacturing 球端铣削的高效刀具路径规划方法，实现高质量制造

IF 9.1 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Robotics and Computer-integrated Manufacturing

Pub Date : 2024-11-26 DOI: 10.1016/j.rcim.2024.102905

Hong-Yu Ma , Yi-Bo Kou , Li-Yong Shen , Chun-Ming Yuan

Triangular mesh representation is extensively utilized in geometric design and reverse engineering. However, in the realm of high quality CNC machining, there is a notable transition from mesh to continuous surface representation for workpieces. This paper presents a novel approach to address this shift, proposing a high-precision and efficient path generation method of ball-end milling specifically designed for triangular meshes. The method integrates considerate surface fitting techniques with productive path planning strategies to optimize machining processes. The method first introduces GNURBS surface fitting adapted for CAM with normal vectors and sharp features preserving, then provides a surface segmentation strategy better suited for machining based on a weighted graph analysis, and finally presents a Fermat spirals path generation scheme with single start and end points. Experimental results and case studies are provided to illustrate and clarify our method. The results show the superior performance and effectiveness of our method concerning surface quality, sharp features, and machining time.

三角网格表示法广泛应用于几何设计和逆向工程。然而，在高质量数控加工领域，工件的表示方法正从网格向连续曲面过渡。本文针对这一转变提出了一种新方法，即专门为三角形网格设计的高精度、高效球端铣削路径生成方法。该方法集成了周到的表面拟合技术和生产路径规划策略，以优化加工过程。该方法首先引入了适用于 CAM 的 GNURBS 表面拟合，并保留了法向量和尖锐特征，然后提供了一种基于加权图分析的更适合加工的表面分割策略，最后提出了一种具有单一起点和终点的费马螺旋路径生成方案。实验结果和案例研究说明并阐明了我们的方法。实验结果表明，我们的方法在表面质量、锐利特征和加工时间方面性能优越，效果显著。

{"title":"Efficient tool path planning method of ball-end milling for high quality manufacturing","authors":"Hong-Yu Ma , Yi-Bo Kou , Li-Yong Shen , Chun-Ming Yuan","doi":"10.1016/j.rcim.2024.102905","DOIUrl":"10.1016/j.rcim.2024.102905","url":null,"abstract":"<div><div>Triangular mesh representation is extensively utilized in geometric design and reverse engineering. However, in the realm of high quality CNC machining, there is a notable transition from mesh to continuous surface representation for workpieces. This paper presents a novel approach to address this shift, proposing a high-precision and efficient path generation method of ball-end milling specifically designed for triangular meshes. The method integrates considerate surface fitting techniques with productive path planning strategies to optimize machining processes. The method first introduces GNURBS surface fitting adapted for CAM with normal vectors and sharp features preserving, then provides a surface segmentation strategy better suited for machining based on a weighted graph analysis, and finally presents a Fermat spirals path generation scheme with single start and end points. Experimental results and case studies are provided to illustrate and clarify our method. The results show the superior performance and effectiveness of our method concerning surface quality, sharp features, and machining time.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"93 ","pages":"Article 102905"},"PeriodicalIF":9.1,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142719940","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A safety posture field framework for mobile manipulators based on human–robot interaction trend and platform-arm coupling motion 基于人机交互趋势和平台-机械臂耦合运动的移动机械手安全姿态场框架

IF 9.1 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Robotics and Computer-integrated Manufacturing

Pub Date : 2024-11-24 DOI: 10.1016/j.rcim.2024.102903

Yong Tao , Jiahao Wan , Yian Song , Xingyu Li , Baicun Wang , Tianmiao Wang , Yiru Wang

Mobile manipulators are increasingly deployed in industrial settings, such as material handling and workpiece loading, where they must safely interact with humans while efficiently completing tasks. Existing motion planning methods for mobile manipulators often struggle to ensure both safety and efficiency in dynamic human-robot interaction environments. This paper proposes a Safety Posture Field framework that addresses these limitations by firstly predicting human motion trends using the improved Long Short-Term Memory neural network and applying these predictions to potential field calculations for both the mobile platform and the robotic arm. During different stages of human-robot interaction, the mobile manipulator places varying emphasis on safety and efficiency while in motion. Additionally, when the robotic arm executes operations, a platform-arm coupling motion strategy is introduced when the potential field detects risks of singularity or local optima, preventing the robotic arm from becoming unstable or failing to reach the target pose in time. This strategy enhances the system's flexibility and operational stability. Comparative experiments in simulation and real-world settings confirm the ability of the framework to maintain high safety standards while improving task efficiency, making it suitable for industrial Human-Robot Interaction applications.

移动机械手越来越多地应用于材料处理和工件装载等工业环境中，它们必须在高效完成任务的同时安全地与人类进行交互。现有的移动机械手运动规划方法往往难以确保动态人机交互环境中的安全和效率。本文提出了一个安全姿态场框架，通过首先使用改进的长短期记忆神经网络预测人类运动趋势，并将这些预测应用于移动平台和机械臂的势场计算，来解决这些局限性。在人机交互的不同阶段，移动操纵器对运动时的安全性和效率的重视程度各不相同。此外，在机械臂执行操作时，当势场检测到奇点或局部最优风险时，会引入平台-机械臂耦合运动策略，防止机械臂变得不稳定或无法及时到达目标姿势。这一策略增强了系统的灵活性和运行稳定性。在模拟和实际环境中进行的对比实验证实，该框架能够在提高任务效率的同时保持较高的安全标准，因此适用于工业领域的人机交互应用。

{"title":"A safety posture field framework for mobile manipulators based on human–robot interaction trend and platform-arm coupling motion","authors":"Yong Tao , Jiahao Wan , Yian Song , Xingyu Li , Baicun Wang , Tianmiao Wang , Yiru Wang","doi":"10.1016/j.rcim.2024.102903","DOIUrl":"10.1016/j.rcim.2024.102903","url":null,"abstract":"<div><div>Mobile manipulators are increasingly deployed in industrial settings, such as material handling and workpiece loading, where they must safely interact with humans while efficiently completing tasks. Existing motion planning methods for mobile manipulators often struggle to ensure both safety and efficiency in dynamic human-robot interaction environments. This paper proposes a Safety Posture Field framework that addresses these limitations by firstly predicting human motion trends using the improved Long Short-Term Memory neural network and applying these predictions to potential field calculations for both the mobile platform and the robotic arm. During different stages of human-robot interaction, the mobile manipulator places varying emphasis on safety and efficiency while in motion. Additionally, when the robotic arm executes operations, a platform-arm coupling motion strategy is introduced when the potential field detects risks of singularity or local optima, preventing the robotic arm from becoming unstable or failing to reach the target pose in time. This strategy enhances the system's flexibility and operational stability. Comparative experiments in simulation and real-world settings confirm the ability of the framework to maintain high safety standards while improving task efficiency, making it suitable for industrial Human-Robot Interaction applications.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"93 ","pages":"Article 102903"},"PeriodicalIF":9.1,"publicationDate":"2024-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142702679","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Processing accuracy improvement of robotic ball-end milling by simultaneously optimizing tool orientation and robotic redundancy 同时优化刀具方向和机器人冗余，提高机器人球端铣削的加工精度

IF 9.1 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Robotics and Computer-integrated Manufacturing

Pub Date : 2024-11-24 DOI: 10.1016/j.rcim.2024.102904

Shizhong Tan, Jixiang Yang, Chengxing Wu, Han Ding

Robotic ball-end milling presents advantages such as a broad workspace, cost-effectiveness, and integration with vision/force sensing, making it a promising method in machinery manufacturing. However, its low stiffness leads to deformation error that seriously affects part profile accuracy. Reducing the deformation error is an effective method to improve the machining accuracy of robotic milling. However, existing research primarily focuses on translational deformation of the robot end effector calculated using average cutting force, overlooking the effect of changes in cutting force and deformation at the tool tip. To address these limitations, an optimization model is proposed to simultaneously optimize tool orientation and redundant angle to minimize force-induced tool tip deformation errors, accounting for cutting force variations at different tool postures. First, an error index for tool tip deformation is introduced, and it considers the comprehensive deformation of the tool tip point instead of the translational deformation of the robot end-effector to offer a more accurate analysis of the machining error. Second, a rapid calculation method for cutter-workpiece engagement is developed, facilitating efficient calculation of cutting forces and enhancing the accuracy of deformation error calculation under various tool orientations. Finally, employing a particle swarm optimization algorithm with multiple constraints, including robot kinematics and tool interference, both tool orientation and robotic redundant angles are optimized to minimize tool error index at each cutter location. Through a comparison test using a simplified aeroengine casing, the proposed method demonstrates effective enhancement of the accuracy of robot milling processing compared with unoptimized and existing studies.

机器人球端铣削具有工作空间宽广、成本效益高、可与视觉/力传感集成等优点，是机械制造领域一种前景广阔的方法。然而，其低刚度会导致变形误差，严重影响零件轮廓精度。减少变形误差是提高机器人铣削加工精度的有效方法。然而，现有的研究主要集中在使用平均切削力计算机器人末端效应器的平移变形，忽略了切削力变化和刀尖变形的影响。为了解决这些局限性，我们提出了一个优化模型，同时优化刀具方向和冗余角度，以最大限度地减少力引起的刀尖变形误差，并考虑到不同刀具姿态下的切削力变化。首先，引入了刀尖变形误差指标，该指标考虑了刀尖点的综合变形，而不是机器人末端执行器的平移变形，从而提供了更精确的加工误差分析。其次，开发了刀具与工件啮合的快速计算方法，便于高效计算切削力，并提高了不同刀具方向下的变形误差计算精度。最后，采用粒子群优化算法，在机器人运动学和刀具干涉等多重约束条件下，对刀具方向和机器人冗余角度进行优化，使每个刀具位置的刀具误差指数最小。通过使用简化的航空发动机机壳进行对比试验，与未优化的方法和现有研究相比，所提出的方法有效提高了机器人铣削加工的精度。

{"title":"Processing accuracy improvement of robotic ball-end milling by simultaneously optimizing tool orientation and robotic redundancy","authors":"Shizhong Tan, Jixiang Yang, Chengxing Wu, Han Ding","doi":"10.1016/j.rcim.2024.102904","DOIUrl":"10.1016/j.rcim.2024.102904","url":null,"abstract":"<div><div>Robotic ball-end milling presents advantages such as a broad workspace, cost-effectiveness, and integration with vision/force sensing, making it a promising method in machinery manufacturing. However, its low stiffness leads to deformation error that seriously affects part profile accuracy. Reducing the deformation error is an effective method to improve the machining accuracy of robotic milling. However, existing research primarily focuses on translational deformation of the robot end effector calculated using average cutting force, overlooking the effect of changes in cutting force and deformation at the tool tip. To address these limitations, an optimization model is proposed to simultaneously optimize tool orientation and redundant angle to minimize force-induced tool tip deformation errors, accounting for cutting force variations at different tool postures. First, an error index for tool tip deformation is introduced, and it considers the comprehensive deformation of the tool tip point instead of the translational deformation of the robot end-effector to offer a more accurate analysis of the machining error. Second, a rapid calculation method for cutter-workpiece engagement is developed, facilitating efficient calculation of cutting forces and enhancing the accuracy of deformation error calculation under various tool orientations. Finally, employing a particle swarm optimization algorithm with multiple constraints, including robot kinematics and tool interference, both tool orientation and robotic redundant angles are optimized to minimize tool error index at each cutter location. Through a comparison test using a simplified aeroengine casing, the proposed method demonstrates effective enhancement of the accuracy of robot milling processing compared with unoptimized and existing studies.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"93 ","pages":"Article 102904"},"PeriodicalIF":9.1,"publicationDate":"2024-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142702827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Knowledge extraction for additive manufacturing process via named entity recognition with LLMs 利用 LLMs 进行命名实体识别，提取增材制造工艺知识

IF 9.1 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Robotics and Computer-integrated Manufacturing

Pub Date : 2024-11-21 DOI: 10.1016/j.rcim.2024.102900

Xuan Liu , John Ahmet Erkoyuncu , Jerry Ying Hsi Fuh , Wen Feng Lu , Bingbing Li

This paper proposes a novel NER framework, leveraging the advanced capabilities of Large Language Models (LLMs), to address the limitations of manually defined taxonomy. Our framework integrates the expert knowledge internalized in both academic materials and LLMs through retrieval-augmented generation (RAG) to automatically customize taxonomies for specific manufacturing processes and adopts two distinct strategies of using LLMs — In-Context Learning (ICL) and fine-tuning to complete manufacturing NER tasks with minimal training data. We demonstrate the framework efficiency through its superior ability to define precise taxonomies, identify and classify process-level entities related to the most popular additive manufacturing process fused deposition modeling (FDM) as case study, achieving a high F1 score of 0.9192.

本文提出了一种新颖的 NER 框架，利用大型语言模型 (LLM) 的先进功能来解决人工定义分类法的局限性。我们的框架通过检索增强生成（RAG）整合了学术材料和 LLM 中内化的专家知识，为特定的制造流程自动定制分类标准，并采用两种不同的 LLM 使用策略--上下文学习（ICL）和微调，以最少的训练数据完成制造 NER 任务。我们以最流行的增材制造工艺熔融沉积建模（FDM）为案例，展示了该框架在定义精确分类标准、识别和分类与该工艺相关的工艺级实体方面的卓越能力，并取得了 0.9192 的高 F1 分数，从而证明了该框架的高效性。

引用次数: 0

Digital Twin-driven multi-scale characterization of machining quality: current status, challenges, and future perspectives 数字孪生驱动的加工质量多尺度表征：现状、挑战和未来展望

IF 9.1 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Robotics and Computer-integrated Manufacturing

Pub Date : 2024-11-21 DOI: 10.1016/j.rcim.2024.102902

Xiangfu Fu , Shuo Li , Hongze Song , Yuqian Lu

The evolution of manufacturing towards intelligent and digital processes requires innovation in machining quality control. While current research primarily addresses single-scale quality control, it overlooks comprehensive multi-scale product quality characterization. Digital twin technology emerges as a potential solution. This review examines digital twin applications in machining quality control, highlighting limitations of traditional methods and exploring multi-scale quality characterization at macro, meso, and micro levels. It evaluates multi-scale quality changes during processing and summarizes comprehensive characterization methods across scales. The study concludes by discussing future prospects for digital twin technology in multi-scale machining quality control and optimization.

制造业向智能化和数字化流程发展，需要在加工质量控制方面进行创新。目前的研究主要针对单一尺度的质量控制，而忽略了全面的多尺度产品质量表征。数字孪生技术是一种潜在的解决方案。本综述探讨了数字孪生技术在机械加工质量控制中的应用，强调了传统方法的局限性，并探索了宏观、中观和微观层面的多尺度质量表征。它评估了加工过程中的多尺度质量变化，并总结了跨尺度的综合表征方法。研究最后讨论了数字孪生技术在多尺度加工质量控制和优化方面的未来前景。

引用次数: 0

A dual knowledge embedded hybrid model based on augmented data and improved loss function for tool wear monitoring 基于增强数据和改进损失函数的双知识嵌入式混合模型，用于刀具磨损监测

IF 9.1 1区计算机科学 Q1 COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS

Robotics and Computer-integrated Manufacturing

Pub Date : 2024-11-15 DOI: 10.1016/j.rcim.2024.102901

Xiaohui Fang , Qinghua Song , Jing Qin , Zhenyang Li , Haifeng Ma , Zhanqiang Liu

Tool wear monitoring (TWM) is essential for enhancing the machining accuracy of intelligent manufacturing systems and ensuring the consistency and reliability of products. The complex and dynamic processing environment demands higher real-time monitoring and generalization ability of TWM. Traditional data-driven models lack guided training in physical processes and are limited by the amount of samples with wear labels. To guide the model to capture the underlying physical mechanism and enhance compliance with the law of tool wear, a dual knowledge embedded hybrid model based on augmented data and improved loss function for TWM is proposed in this paper. The second training data source is obtained by constructing the mapping relationship between cutting force and tool wear, which effectively complements and enhances the physical characteristics between the data and addresses the issue of insufficient labeled data in actual network training. Subsequently, a structure integrating serial convolution, parallel convolution, bidirectional gated recurrent unit (BiGRU) and attention mechanism is developed to extract the spatial and temporal features in time series data. Moreover, Based on the physical law of tool wear, an improved loss function with physical constraints is proposed to improve the physical consistency of the model. The experimental results show that the model prediction RMSE error is reduced by 12.67% after augmented data compared to a single data source, and the RMSE error of the prediction result is reduced by 25.16% at most after the improvement of the loss function. The model has high prediction accuracy within short training epochs and good real-time performance. The proposed approach provides a modeling strategy with low computational resource requirements based on the fusion of physical and data information.

刀具磨损监测（TWM）对于提高智能制造系统的加工精度、确保产品的一致性和可靠性至关重要。复杂多变的加工环境要求 TWM 具有更高的实时监控和概括能力。传统的数据驱动模型缺乏对物理过程的指导性训练，并且受到带有磨损标签的样本量的限制。为了引导模型捕捉潜在的物理机制并增强对刀具磨损规律的遵从，本文提出了一种基于增强数据和改进损失函数的双知识嵌入式混合模型，用于 TWM。通过构建切削力与刀具磨损之间的映射关系获得第二训练数据源，有效补充和增强了数据之间的物理特性，解决了实际网络训练中标注数据不足的问题。随后，开发了集串行卷积、并行卷积、双向门控递归单元（BiGRU）和注意力机制于一体的结构，以提取时间序列数据的时空特征。此外，根据刀具磨损的物理规律，提出了一种带有物理约束的改进损失函数，以提高模型的物理一致性。实验结果表明，与单一数据源相比，增强数据后的模型预测均方根误差降低了 12.67%，损失函数改进后的预测结果均方根误差最多降低了 25.16%。该模型在较短的训练历时内具有较高的预测精度和良好的实时性。所提出的方法提供了一种基于物理信息和数据信息融合的低计算资源需求建模策略。

{"title":"A dual knowledge embedded hybrid model based on augmented data and improved loss function for tool wear monitoring","authors":"Xiaohui Fang , Qinghua Song , Jing Qin , Zhenyang Li , Haifeng Ma , Zhanqiang Liu","doi":"10.1016/j.rcim.2024.102901","DOIUrl":"10.1016/j.rcim.2024.102901","url":null,"abstract":"<div><div>Tool wear monitoring (TWM) is essential for enhancing the machining accuracy of intelligent manufacturing systems and ensuring the consistency and reliability of products. The complex and dynamic processing environment demands higher real-time monitoring and generalization ability of TWM. Traditional data-driven models lack guided training in physical processes and are limited by the amount of samples with wear labels. To guide the model to capture the underlying physical mechanism and enhance compliance with the law of tool wear, a dual knowledge embedded hybrid model based on augmented data and improved loss function for TWM is proposed in this paper. The second training data source is obtained by constructing the mapping relationship between cutting force and tool wear, which effectively complements and enhances the physical characteristics between the data and addresses the issue of insufficient labeled data in actual network training. Subsequently, a structure integrating serial convolution, parallel convolution, bidirectional gated recurrent unit (BiGRU) and attention mechanism is developed to extract the spatial and temporal features in time series data. Moreover, Based on the physical law of tool wear, an improved loss function with physical constraints is proposed to improve the physical consistency of the model. The experimental results show that the model prediction RMSE error is reduced by 12.67% after augmented data compared to a single data source, and the RMSE error of the prediction result is reduced by 25.16% at most after the improvement of the loss function. The model has high prediction accuracy within short training epochs and good real-time performance. The proposed approach provides a modeling strategy with low computational resource requirements based on the fusion of physical and data information.</div></div>","PeriodicalId":21452,"journal":{"name":"Robotics and Computer-integrated Manufacturing","volume":"92 ","pages":"Article 102901"},"PeriodicalIF":9.1,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142637642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0