SBIR Phase I: Comprehensive, Human-Centered, Safety System Using Physiological and Behavioral Sensing to Predict and Prevent Workplace Accidents

小型企业创新研究(SBIR)第一阶段项目的更广泛/商业影响是通过使用可穿戴技术识别和预测事故，更好地保护工人免受工作场所的危险。与人为因素有关的事故占伤害的80%，目前可用的安全产品没有解决这一问题。该解决方案利用可穿戴技术自动收集员工的生理和行为数据。这些数据被纳入机器学习模型，以识别安全事件和险些发生的事故。这种针对工人安全的创新方法通过使用机器学习来解释工人的生理和行为产生的信号，从而增强了对科学和技术的理解。对工作场所危险的反应被用来触发警报，以预测和预防工作场所事故。这个安全系统为预测事故可能性的机器学习模型提供了基础，这样安全人员就可以在工人受伤之前进行干预。该项目的目标是防止受伤，拯救生命，并使公司能够节省保险成本、债务和工作损失时间。这个SBIR第一阶段项目旨在开发一种使用人体内置传感器来识别安全危险的安全系统。通过使用可穿戴技术自动连续收集实时生理和行为数据，将开发机器学习模型来识别安全事件，从而能够预测和预防事故。这项研究的智力价值在于：1)验证人类对滑倒和绊倒的反应是否类似、可测量；2)开发机器学习模型以准确识别滑倒和绊倒及其强度；3)开发机器学习模型以评估未来安全事故的风险；以及4)验证数据可以通过整个工作流程进行处理，向工人和安全人员提供实时警报。数据将从使用研究级可穿戴设备滑倒和绊倒的人类受试者那里收集。这项研究的预期结果将为安全系统提供基础，该系统用于触发安全警报并确定风险级别，以拯救生命并防止与滑倒和绊倒相关的事故。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader/commercial impact of this Small Business Innovation Research (SBIR) Phase I project is to better protect workers from hazards in the workplace through the use of wearable technology to identify and predict accidents. Human-factor related accidents account for 80% of injuries and are not being addressed with currently available safety products. This solution utilizes wearable technology to automate the collection of physiological and behavioral data from workers. The data is incorporated into machine learning models to identify safety incidents and near-misses. This innovative approach to worker safety enhances scientific and technological understanding by using machine learning to interpret signals generated by a worker’s physiology and behaviors. Responses to hazards in the workplace are used to trigger alerts that predict and prevent workplace accidents. This safety system provides the basis for machine learning models that predict the likelihood of accidents so safety personnel can intervene before the worker is injured. The goal of this project is to prevent injuries, save lives, and enable companies to realize savings in insurance costs, liabilities, and lost time from the job.This SBIR Phase I project aims to develop a safety system that uses the human body’s built-in sensors to identify safety hazards. By automating the continuous collection of real-time physiological and behavioral data using wearable technology, machine learning models will be developed to identify safety incidents, enabling the prediction and prevention of accidents. The intellectual merit of the research is to: 1) verify that humans respond in similar, measurable ways to slips and trips, 2) develop machine learning models to accurately identify slips and trips and their intensity, 3) develop machine learning models to assess the risk of future safety accidents, and 4) verify that data can be processed through the entire workflow to provide real-time alerts to the worker and safety personnel. Data will be collected from human subjects subjected to slips and trips using research-grade wearables. The anticipated output of this research will provide the basis for a safety system used to trigger safety alerts and identify risk levels to save lives and prevent accidents related to slips and trips.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.