Anind K. Dey is an Associate Professor in the Human-Computer Interaction Institute at Carnegie Mellon University. He is the director of the Ubicomp Lab, which performs research at the intersection of ubiquitous computing, human-computer interaction and machine learning, in the areas of mobile computing, health and sustainability among others. He has authored over 100 papers on these topics and serves on the editorial board of several journals. Anind received his PhD in computer science from Georgia Tech, along with a Masters of Science in both Computer Science and Aerospace Engineering. He received his Bachelors of Applied Science in Computer Engineering from Simon Fraser University.
**Understanding Human Behavior Through Context**
Commodity smart phones have made the visions of ubiquitous computing and context-awareness common place. We call these phones "smart phones" simply because they have a mobile operating system, not because they are smart. In fact, they are pretty dumb. They know nothing about their users, despite the fact that they spend hours a day with them. We have been using these phones to collect a wide variety of data to enable a wide variety of context-aware user experiences, focusing on experiences that require a truly "smart" phone. In this talk, I will describe the requirements and a vision for making a smart phone, and break a number of assumptions about phone usage that will impact the way we design context-enabled smart phone applications.
Dr. Raja Sengupta is currently Professor and Program Leader of the Systems program in Civil & Environmental Engineering at the University of California, Berkeley. He received his Ph.d from the EECS department of the University of Michigan, at Ann Arbor. His current research interests are in smart cities, mobile computing, cloud computing, human behavior change technology, transportation demand, robotics, drones, and control theory. He was Program Chair of the IEEE Conference on Autonomous Intelligent Networked Systems 2003 and Co-General Chair of the first ACM MOBICOM Workshop on Vehicular Ad-hoc Networks held in 2004, Co-Chair of the Program Committee for the second ACM MOBICOM Workshop on Vehicular Ad-hoc Networks held in 2005, Program Chair for the First International Symposium on Vehicular Computing Systems 2008, and Co-General Chair of IEEE WIVEC 2011. He served on the editorial boards of the IEEE Control Systems Magazine and the Journal of Intelligent Transportation Systems. He was recognized by USDOT in 2012 with the Connected Vehicle Technology Award.
**Bigraphical Representation of Context: The BigActor Model**
We discuss the representation of context in Milner's Bigraphical Model. Computation has the semantics of the Actor model, and the two together create a hybrid called the BigActor Model. The model has an operational semantics that mirrors feedback control i.e,. computation observes the context, computes, and may then actuate it. The model reveals a basic correctness property required of context-aware programs, and we present sufficient conditions that are practical and correct. The model is implemented as a library for aerobots, and we discuss a case study executed using the implementation in the Algarve with the Air Force of Portugal. The case study tests the a drone embedded in a distributed system with drifters and satellites. The case study scenario emulates illegal bilge dumping by a ship on the high seas, and the response by maritime forces to the event.