Coordinated Supervisory Control System for Smart Manufacturing

智能制造协调监控系统

• 批准号: 1853454
• 负责人: Qing Chang
• 金额: $ 49.94万
• 依托单位: University of Virginia Main Campus
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-09-15 至 2024-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1853454&HistoricalAwards=false
• 关键词: Coordinated; Supervisory; Control; System; Smart

This project will support fundamental research in smart manufacturing processes and systems, promoting the progress of science as well as national prosperity. Recent advancements in sensors, information, communication, and robotics technologies have driven better ways of doing manufacturing. Modern manufacturing systems are equipped with multiple sensors, communication devices, and material handling robots that operate alongside other autonomous machines and human coworkers/supervisors. However, the full potential of these technological advances in improving production efficiency and quality has not been realized. This is because manufacturing systems are inherently stochastic and nonlinear, and there is a lack of theoretical and technical understanding of real-time model-based prediction of manufacturing performance and production control. This research will establish novel data-enabled models for predicting manufacturing system performance under uncertainty and adaptive, robust, and scalable control and coordination algorithms for production control. The new knowledge will provide manufacturers with a rigorous quantitative tool for real-time monitoring and control of the complex manufacturing systems. The results of this research will contribute to both the theory and the practice of smart manufacturing and will be very useful to the entire U.S. manufacturing sector for maximizing productivity and economic benefit. This research incorporates goals with industrial needs, helps broaden the participation of underrepresented groups in research, and enhances engineering education. This research will fill the knowledge gap via the integrated study of production dynamics and multiple robots material dispatching decisions to address their interactions in real-time, which is significant for manufacturing systems efficiency. The research approach will be a novel combination of dynamic systems modeling of material flow through production processes along with dynamic multirobot task assignment under uncertainty for optimizing production performance metrics. The research team aims to: a) establish a data-enabled mathematical framework to describe real-time manufacturing systems with dynamic production and multirobot operations to enhance the understanding of dynamic manufacturing processes and systems; and b) establish the scientific and technological foundation in adaptive control and decentralized decision making for multirobot system working with machines/humans. The Coordinated Supervisory Control System is based on the holistic view of physical system analysis, advanced data-driven modeling, and adaptive control. The methodology is transformable to other dynamic systems with distributed sensors and data, such as transportation, supply chain, and health care management.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

该项目将支持智能制造工艺和系统的基础研究，促进科学进步和国家繁荣。最近传感器、信息、通信和机器人技术的进步推动了更好的制造方式。现代制造系统配备了多个传感器、通信设备和物料搬运机器人，这些机器人与其他自主机器和人类同事/主管一起操作。然而，这些技术进步在提高生产效率和质量方面的潜力尚未充分发挥出来。这是因为制造系统本质上是随机和非线性的，并且缺乏对基于实时模型的制造性能预测和生产控制的理论和技术理解。本研究将建立新的数据支持模型，用于预测不确定性下的制造系统性能，以及用于生产控制的自适应、鲁棒和可扩展控制和协调算法。新知识将为制造商提供严格的定量工具，用于实时监测和控制复杂的制造系统。本研究的结果将对智能制造的理论和实践做出贡献，并将对整个美国制造业的生产力和经济效益最大化非常有用。这项研究将目标与工业需求结合起来，有助于扩大代表性不足的群体在研究中的参与，并加强工程教育。本研究将通过对生产动力学和多机器人材料调度决策的集成研究来填补知识空白，以解决它们之间的实时交互，这对制造系统的效率具有重要意义。研究方法将是生产过程中物料流动的动态系统建模与不确定性下动态多机器人任务分配的新组合，以优化生产绩效指标。研究团队的目标是：a)建立一个数据支持的数学框架来描述具有动态生产和多机器人操作的实时制造系统，以增强对动态制造过程和系统的理解；b)为机器/人协同工作的多机器人系统的自适应控制和分散决策建立科学技术基础。协调监控系统基于物理系统分析的整体视图，先进的数据驱动建模和自适应控制。该方法可转换为具有分布式传感器和数据的其他动态系统，例如运输，供应链和医疗保健管理。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Integrated process-system modelling and control through graph neural network and reinforcement learning

通过图神经网络和强化学习集成过程系统建模和控制

• DOI: 10.1016/j.cirp.2021.04.056
• 发表时间: 2021
• 期刊: CIRP Annals
• 作者: Huang, Jing;Zhang, Jianjing;Chang, Qing;Gao, Robert X.
• 通讯作者: Gao, Robert X.

Dynamic modeling and analysis of multi-product flexible production line

• DOI: 10.1080/0951192x.2023.2177746
• 发表时间: 2023-02
• 期刊: International Journal of Computer Integrated Manufacturing
• 影响因子: 4.1
• 作者: M. Waseem;Chen Li;Qing Chang

Hybrid feedback and reinforcement learning-based control of machine cycle time for a multi-stage production system

• DOI: 10.1016/j.jmsy.2022.09.020
• 发表时间: 2022-10
• 期刊: Journal of Manufacturing Systems
• 影响因子: 12.1
• 作者: Chen Li;Qing Chang

Integrated Process-System Modeling and Performance Analysis for Serial Production Lines

• DOI: 10.1109/lra.2022.3181741
• 发表时间: 2022
• 期刊: IEEE Robotics and Automation Letters
• 影响因子: 5.2
• 作者: Chen Li;Q. Chang;G. Xiao;J. Arinez

User-guided motion planning with reinforcement learning for human-robot collaboration in smart manufacturing

智能制造中人机协作的用户引导运动规划和强化学习

• DOI: 10.1016/j.eswa.2022.118291
• 发表时间: 2022
• 期刊: Expert Systems with Applications
• 影响因子: 8.5
• 作者: Yu, Tian;Chang, Qing
• 通讯作者: Chang, Qing