CPS: Synergy: Collaborative Research: Cyber-Physical Sensing, Modeling, and Control with Augmented Reality for Smart Manufacturing Workforce Training and Operations Management

CPS：协同：协作研究：用于智能制造劳动力培训和运营管理的网络物理传感、建模和增强现实控制

• 批准号: 1646065
• 负责人: Zhihai He
• 金额: $ 19.6万
• 依托单位: University of Missouri-Columbia
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-02-01 至 2021-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1646065&HistoricalAwards=false
• 关键词: CPS; Synergy; Collaborative; Research; Cyber

Smart manufacturing integrates information, technology, and human ingenuity to inspire the next revolution in the manufacturing industry. Manufacturing has been identified as a key strategic investment area by the U.S. government, private sector, and university leaders to spur innovation and keep America competitive. However, the lack of new methodologies and tools is challenging continuous innovation in the smart manufacturing industry. This award supports fundamental research to develop a cyber-physical sensing, modeling, and control infrastructure, coupled with augmented reality, to significantly improve the efficiency of future workforce training, performance of operations management, safety and comfort of workers for smart manufacturing. Results from this research are expected to transform the practice of worker-machine-task coordination and provide a powerful tool for operations management. This research involves several disciplines including sensing, data analytics, modeling, control, augmented reality, and workforce training and will provide unique interdisciplinary training opportunities for students and future manufacturing engineers. An effective way for manufacturers to tackle and outpace the increasing complexity of product designs and ever-shortening product lifecycles is to effectively develop and assist the workforce. Yet the current management of manufacturing workforce systems relies mostly on the traditional methods of data collection and modeling, such as subjective observations and after-the-fact statistics of workforce performance, which has reached a bottleneck in effectiveness. The goal of this project is to investigate an integrated set of cyber-physical system methods and tools to sense, understand, characterize, model, and optimize the learning and operation of manufacturing workers, so as to achieve significantly improved efficiency in worker training, effectiveness of behavioral operations management, and safety of front-line workers. The research team will instrument a suite of sensors to gather real-time data about individual workers, worker-machine interactions, and the working environment,develop advanced methods and tools to track and understand workers' actions and physiological status, and detect their knowledge and skill deficiencies or assistance needs in real time. The project will also establish mathematical models that encode the manufacturing process in the research sensing and analysis framework, characterize the efficiency of worker-machine-task coordination, model the learning curves of individual workers, investigate various multi-modal augmented reality-based visualization, guidance, control, and intervention schemes to improve task efficiency and worker safety, and deploy, test, and conduct comprehensive performance assessments of the Researched technologies.

智能制造融合了信息、技术和人类的聪明才智，激发了制造业的下一次革命。制造业已被美国政府、私营部门和大学领导人确定为一个关键的战略投资领域，以刺激创新并保持美国的竞争力。然而，缺乏新的方法和工具正在挑战智能制造行业的持续创新。该奖项支持基础研究，以开发网络物理传感，建模和控制基础设施，再加上增强现实，以显着提高未来劳动力培训的效率，运营管理的性能，工人的安全性和舒适性，以实现智能制造。本研究的结果可望改变人-机-任务协调的实践，并为作业管理提供一个强有力的工具。这项研究涉及多个学科，包括传感，数据分析，建模，控制，增强现实和劳动力培训，并将为学生和未来的制造工程师提供独特的跨学科培训机会。制造商应对产品设计日益复杂和产品生命周期不断缩短的有效方法是有效地开发和协助员工。然而，目前的制造业劳动力系统的管理大多依赖于传统的数据收集和建模方法，如主观观察和事后统计的劳动力绩效，这已经达到了有效性的瓶颈。该项目的目标是研究一套集成的网络物理系统方法和工具，以感知、理解、表征、建模和优化制造工人的学习和操作，从而显著提高工人培训的效率、行为操作管理的有效性和一线工人的安全性。该研究团队将配备一套传感器，以收集有关单个工人，工人-机器交互和工作环境的实时数据，开发先进的方法和工具来跟踪和了解工人的行动和生理状态，并真实的时间检测他们的知识和技能不足或援助需求。该项目还将建立数学模型，在研究传感和分析框架中对制造过程进行编码，表征工人-机器-任务协调的效率，对单个工人的学习曲线进行建模，研究各种基于多模态增强现实的可视化，指导，控制和干预方案，以提高任务效率和工人安全，并部署，测试，并对所研究的技术进行全面的性能评估。

项目成果

Pedestrian Motion Trajectory Prediction With Stereo-Based 3D Deep Pose Estimation and Trajectory Learning

利用基于立体的 3D 深度姿势估计和轨迹学习进行行人运动轨迹预测

• 发表时间: 2020
• 期刊: IEEE access
• 影响因子: 3.9
• 作者: Jianqi Zhong, Hao Sun