Physical Knowledge Discovery from "Dirty" Smart City Data
Prof. Pei ZHANG Associate Research Professor, Electrical and Computer Engineering, College of Engineering, Carnegie Mellon University
Abstract
In many smart city applications, direct sensing of desired events is difficult and often impossible. This is often due to deployment difficulties, lack of available sensors, and cost of maintenance especially across an entire city. This talk will examine indirect sensing framework that infers information from physically sensed information. This approach incorporates physical modeling and information to reduces the number of sensors, improves accuracy and decrease the need for labeled training data. The talk explores this framework using real deployments are examples. 1) vehicular sensing, where the system can infer air pollution and actuate base on existing air and mobility models as well as sensed data. 2) Infrastructure-based sensors that uses vibration sensors to infer occupant information including location, identity, and status. Through these projects, the talk will explore ways in incorporating physical and heuristic information into data models and their tradeoffs.
Biography
Pei Zhang is an associate research professor in the ECE departments at Carnegie Mellon University. He received his bachelor's degree with honors from California Institute of Technology in 2002, and his Ph.D. degree in Electrical Engineering from Princeton University in 2008. While at Princeton University, he developed the ZebraNet system, which is used to track zebras in Kenya. It was the first deployed, wireless, ad- hoc, mobile sensor network. His recent work includes SensorFly (focus on groups of autonomous miniature-helicopter based sensor nodes) and MARS (Muscle Activity Recognition). Beyond research publications, his work has been featured in popular media including CNN, Science Channel, Discovery Channel, CBS News, CNET, Popular Science, BBC Focus, etc. He is also a co-founder of the startup Vibradotech. In addition, he has won several awards including the NSF CAREER award, SenSys Test of Time Award, Google faculty award, and a member of the Department of Defense Computer Science Studies Panel.