Current Institution: Stony Brook University
Bio: Yu Li is a Ph.D. candidate in Civil Engineering at Stony Brook University. She obtained her M.S. degree in Systems Engineering from Missouri University of Science and Technology in 2019 and her B.S. degree in Engineering Management from Yichun University, China, in 2012. Her research focuses on sensing and machine learning methods for building human Cyber-Physical Systems (hCPS) to address human-machine interaction challenges for complex civil infrastructure, such as the bridge inspection application. Her Ph.D. dissertation research will introduce lightweight, user-friendly human sensing methods to the workplace, reliable deep learning models for understanding the heterogeneous, evolving workforce, and on-the-job training and assistance methods to workers in human-machine interaction.
Abstract: A Cyber-Physical System for collaborative human-robot inspection of civil infrastructure
Bridges need to be inspected every two years to avoid catastrophic incidents. Robots are being adopted in bridge inspection. To improve worker-robot collaboration, I propose a human cyber-physical system (hCPS) for bridge inspection, including a VR-based training and assessment module named TASBID and an AI-enabled module for inspection assistance. TASBID is designed for workers to practice their skills, develop their confidence, and collaborate with robots. I also design an individualized deep learning architecture to address the challenge of worker heterogeneity posed by worker-robot interaction. Consequently, robots can reliably understand every worker’s intention. A lightweight sensing system could be achieved by developing a teacher-student network to make an AI-enabled system implementable in the bridge site. The proposed hCPS can analyze and identify if a worker lacks certain cognitive, psychomotor, or sensory ability in performing any specific task and recommend a customized training plan to improve worker-robot collaboration.