research

Traditional apprenticeship programs do not provide sufficient feedback on postures and motions to inform trainees of their functional adaptation. In other fields (e.g., athletics), novices are encouraged to learn the intricacies of effective movement techniques through professional instruction. This study is to investigate how craft workers may learn and practice proper working techniques to intuitively understand how to move safely and efficiently. Analyzing motion data with AI can articulate experts' 'physical wisdom'. With emerging technology, we might ask whether we can convey the 'physical wisdom' as functional learning to train apprentices. With a combination of an IMC system and mixed reality, users can observe and follow 3D computer graphic (CG) animations of experts' motion via stereoscopic displays generated from captured data. Furthermore, by combining the system with digital information from the actual workspaces, workers will be able to improve their efficiency and safety at complex and multi-scale worksites.

Development of real-time, on-site ergonomic assessment systems that utilize wearable sensors and computer vision to monitor worker postures, movements, and workload. These systems provide immediate feedback to workers and supervisors, enabling proactive intervention to prevent work-related musculoskeletal disorders. Our approach integrates multiple data streams including IMU sensors, surface EMG, and video analysis to create comprehensive workload profiles that inform evidence-based ergonomic redesigns and intervention strategies.

Advanced biomechanical modeling and analysis to identify injury risk factors in occupational settings. We develop computational models that predict joint forces, muscle activation patterns, and cumulative exposure to risk factors. This research combines laboratory-based motion capture studies with field observations to validate practical assessment tools for use in real workplace environments. Our work supports the design of targeted interventions that reduce physical demands while maintaining productivity.

Investigation of cognitive demands in complex work environments and their impact on human performance and safety. We employ multi-modal assessment methods including subjective ratings, physiological measures, and performance metrics to characterize mental workload. This research informs the design of work systems that optimize cognitive resources, reduce mental fatigue, and support sustainable careers. Applications include manufacturing, healthcare, and other safety-critical domains where cognitive demands are high.

Systematic development and evaluation of ergonomic interventions to reduce workplace injury risk and enhance worker well-being. Our approach combines participatory design methods with rigorous evaluation protocols to ensure interventions are both effective and acceptable to workers. We study engineering controls, administrative strategies, work-rest scheduling, and behavioral interventions. Research outcomes include validated intervention toolkits and implementation guidelines for practitioners.

Development and validation of wearable sensor systems for continuous monitoring of worker exposure to ergonomic risk factors. We integrate IMU sensors, force sensors, and physiological monitors to capture comprehensive data on physical demands during actual work. Machine learning algorithms process sensor data to automatically classify tasks, detect hazardous postures, and quantify cumulative exposure. This research enables scalable, objective ergonomic assessments that support data-driven workplace improvements.