PFI-TT: Smart pose and position guidance system for construction workers' safety and productivity

PFI-TT：智能姿势和位置引导系统，确保建筑工人的安全和生产力

• 批准号: 1919068
• 负责人: Yong Cho
• 金额: $ 25万
• 依托单位: Georgia Tech Research Corporation
• 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: 2019-08-01 至 2024-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1919068&HistoricalAwards=false
• 关键词: PFI; TT; Smart; pose; position

The broader impact/commercial potential of this Partnerships for Innovation - Technology Translation (PFI-TT) project is to improve worker safety and productivity in dynamic construction job sites by enabling commercialization of a wireless pose and position guidance system. If successful, this project is expected to radically change job hazard safety management for the $1.2 trillion U.S. construction industry in which approximately three to four workers die every working day. The proposed technology is expected to help owners, contractors, and construction managers to make better and safer decisions for dynamic construction work. In addition, this study can provide valuable data regarding the safety condition of workers, which benefits the individuals' health as well as safety management and society in general. The collected data will preserve safety knowledge within the company and help knowledge transfer from experienced professionals to new generations. The research project will also provide an opportunity to engage underrepresented students from diverse socio-economic backgrounds to learn about the challenging engineering problems in the construction industry and equip them with scientific knowledge to address those problems.The proposed project aims to provide a holistic solution for intelligently monitoring individual workers' exposure to hazards and unsafe postures. For a system to be feasible as well as practically acceptable in construction which has dynamic and transient work environmental characteristics, the system needs to satisfy many factors: cost, ease of use, scalability, real-time ability, network connectivity, form factor, system deployment, accuracy, reliability, and privacy. While satisfying the factors, a Building Information Model (BIM)-driven Internet of Things (IoT) application will be used as a central platform for user interface and sensor tracking algorithms in the proposed PFI project. In particular, when the construction work schedules are set and incorporated into BIM, daily and weekly work activity information is extracted. These data can serve to automatically or semi-automatically identify potential hazards in advance based on construction progress. In addition to the location-based safety risks, postural working conditions of the workers are monitored through a motion recognition function using machine learning algorithms. The monitored and quantified information can provide valuable data regarding the safety condition of workers, empower our knowledge about each worker's health and safety-related behavior in a quantitative manner, and thus deliver an objective, comprehensive safety reinforcing system to a job site.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.

该创新-技术转化合作伙伴关系（PFI-TT）项目的更广泛影响/商业潜力是通过实现无线姿势和位置引导系统的商业化来提高动态建筑工地中的工人安全和生产力。如果成功，该项目预计将从根本上改变价值1.2万亿美元的美国建筑业的工作危害安全管理，其中每个工作日约有3至4名工人死亡。预计该技术将帮助业主、承包商和施工经理为动态施工工作做出更好、更安全的决策。此外，这项研究可以提供有关工人安全状况的有价值的数据，这有利于个人的健康以及安全管理和整个社会。收集的数据将保存公司内部的安全知识，并帮助经验丰富的专业人员将知识转移给新一代。该研究项目还将提供一个机会，让来自不同社会经济背景的代表性不足的学生了解建筑行业具有挑战性的工程问题，并为他们提供科学知识来解决这些问题。拟议项目旨在提供一个全面的解决方案，智能监控个别工人暴露于危险和不安全的姿势。对于一个系统是可行的，以及实际上可接受的建设，具有动态和瞬态的工作环境的特点，该系统需要满足许多因素：成本，易用性，可扩展性，实时能力，网络连接性，形状因子，系统部署，准确性，可靠性和隐私。在满足这些因素的同时，建筑信息模型（BIM）驱动的物联网（IoT）应用程序将被用作拟议PFI项目中用户界面和传感器跟踪算法的中心平台。特别是，当设置施工工作计划并将其纳入BIM时，会提取每日和每周的工作活动信息。这些数据可用于根据施工进度提前自动或半自动识别潜在危险。除了基于位置的安全风险外，工人的姿势工作条件还通过使用机器学习算法的运动识别功能进行监控。监测和量化的信息可以提供关于工人的安全状况的有价值的数据，以量化的方式增强我们关于每个工人的健康和安全相关行为的知识，从而提供目标，该奖项反映了NSF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进行评估，被认为值得支持审查标准。

项目成果

Effective inertial sensor quantity and locations on a body for deep learning-based worker's motion recognition

• DOI: 10.1016/j.autcon.2020.103126
• 发表时间: 2020-05-01
• 期刊: AUTOMATION IN CONSTRUCTION
• 影响因子: 10.3
• 作者: Kim, Kinam;Cho, Yong K.
• 通讯作者: Cho, Yong K.

Automatic Recognition of Workers’ Motions in Highway Construction by Using Motion Sensors and Long Short-Term Memory Networks

使用运动传感器和长短期记忆网络自动识别公路施工中工人的动作

• DOI: 10.1061/(asce)co.1943-7862.0002001
• 发表时间: 2021
• 期刊: Journal of Construction Engineering and Management
• 影响因子: 5.1
• 作者: Kim, Kinam;Cho, Yong K.
• 通讯作者: Cho, Yong K.

Signal Processing and Alert Logic Evaluation for IoT–Based Work Zone Proximity Safety System

• DOI: 10.1061/jcemd4.coeng-12417
• 发表时间: 2023-02
• 期刊: Journal of Construction Engineering and Management
• 影响因子: 5.1
• 作者: Kinam Kim;Inbae Jeong;Yong K. Cho

Evaluation of Machine Learning Algorithms for Worker’s Motion Recognition Using Motion Sensors

• DOI: 10.1061/9780784482438.007
• 发表时间: 2019-06
• 期刊: Computing in Civil Engineering 2019
• 作者: Kinam Kim;Jingdao Chen;Y. Cho