Structures as Sensors: Using Structures to Indirectly Monitor Humans and Surroundings

This talk introduces “structures as sensors” for the indirect monitoring of cyber-physical-human systems by sensing and analyzing their noisy physical structural responses. Smart structures are designed to sense, understand, and respond to various situations involving the structure itself, the humans within, and the surrounding environment. Traditional direct monitoring approaches using dedicated sensors often result in dense and heterogeneous sensing systems that are difficult to install and maintain in large-scale structures. The conditions of the structure itself, the environment around, and the activities of users within all have a direct impact on the physical responses of the structure. For example, human walking induces building floor vibrations, uneven road surfaces and bridge settlement change vehicle motions, etc.
This talk focuses on my “structures as sensors” approach that utilizes the structure as a sensing medium to indirectly infer multiple types of hidden information relating to the structure (e.g. the users and the surrounding). By using the same set of sensors for multiple types of information, and because of wave propagation characteristics, this approach significantly reduces the number and type of sensors needed to install and maintain. Challenges lie, however, in creating robust inference models for analyzing convoluted noisy structural response data (e.g., a mixture of building responses due to human activities, outside traffic, seismic events). To this end, I developed physics-guided data analytics approaches that combine statistical signal processing and machine learning with physical principles (e.g., wave propagation, human motions, structural dynamics, etc.). Specifically, I present two projects as examples of this approach; 1) Vehicles as Sensors: indirect infrastructure health monitoring through vehicle responses; and 2) Buildings as Sensors: occupant tracking and characterization through footstep-induced building vibrations. I will also present results from the real-world experiments, including our 3-year railway and eldercare center deployments.
 
Bio: Hae Young Noh is an assistant professor in the Dept. of Civil & Environmental Engineering and a courtesy assistant professor in the Dept. of Electrical & Computer Engineering at Carnegie Mellon University. Her research focuses on indirect sensing and physics-guided data analytics to enable low-cost and non-intrusive monitoring of cyber-physical-human systems. The result of her work has been deployed in a number of real-world applications from trains, to the Amish community, to eldercare centers, to pig farms. She received her Ph.D. and M.S. degrees in Civil and Environmental Engineering and the second M.S. degree in Electrical Engineering at Stanford University. She earned her B.S. degree in Mechanical and Aerospace Engineering at Cornell University. She received a number of awards, including the Google Faculty Research Awards in 2013 and 2016, the Dean’s Early Career Fellowship in 2018, and the National Science Foundation CAREER award in 2017.