Mobile Crowd-Sensing

Many of us spend the bulk of our time -- both awake and asleep -- in the company of smartphones, wearable fitness trackers, and countless other devices with sensing and interaction capabilities. Given their powers of observation, one might expect these devices to learn a great deal about our objectives, preferences, routines, and capabilities. One might expect them to accurately predict our behaviors and actively assist us in our daily lives. Such expectations are not being met. Our devices exhibit very few of the intelligent behaviors we might expect a human observer to acquire. They rarely take initiative to assist us. When they do take initiative they often do so at inopportune times, and in cases where the initiated action comes at an opportune moment chances are quite high that the success is coincidental rather than purposeful. To guard against such long odds of successful interaction our devices resort to passive strategies, patiently awaiting our commands while doing a bit of background computation with the hope of being useful when such commands arrive. This is the state-of-the-art in current cyber-human system (CHS) technology. Despite many years of recent research and much confidence in the future of CHSs, our current CHSs remain passively facilitative rather than actively helpful.

We envision an improved state-of-the-art in which computational and human elements are deeply integrated by virtue of mutually understood and coordinated objectives; where this understanding and coordination drives effective sensing and actuation of both computational and human elements; and where such sensing and actuation serve to optimize the allocation of scarce human and computational resources. We are actively working on several topics including device-initiated interruption, human-response validity, device autonomy, and the inference of human objectives and capabilities. We have been developing Sensus as a technological foundation for our research [1]. We seek individuals who wish to pursue a Ph.D. focusing on any of these topics or other related topics. If you are interested, please send your current C.V., publications (if any), and a summary of your interests to



  1. [xiong2016sensus] Xiong, H., Y. Huang, L. Barnes, and M. Gerber, "Sensus: A Cross-Platform, General-Purpose System for Mobile Crowdsensing in Human-Subject Studies", ACM International Joint Conference on Pervasive and Ubiquitous Computing, Heidelberg, Germany, ACM, In Press.