Abstract: This project is developing methods for human context awareness in smart buildings, aging-in-place, smart authorization, and emergency response. We enable this using inference using radio frequency (RF) sensing networks, in which channel measurements are made by deployed wireless networking devices. In RF sensing networks, the network is the sensor. We are developing methods to learn human context for smart facilities and elder care, including:

Abstract: The CYPRESS (CYber Physical RESiliance and Sustainability) project explores techniques for dependability and resilience in instrumented cyberphysical spaces (ICPS) where physical spaces have been instrumented with “intelligence” through heterogeneous sensing, actuation and communication mechanisms.

Abstract: This project addresses architectural considerations in the design and operation of future power grids. This includes the consideration of sophisticated sensing, communication, and actuation capabilities on the system's reliability, price volatility, and economic and environmental efficiency.

Abstract: The objective of this research is to develop foundations for the newly emerging generation of Networked Cyber-Physical Systems (NCPS). NCPS present many intellectual challenges not suitably addressed by existing computing paradigms. They must achieve system-wide objectives through local, asynchronous actions, using distributed control loops through the environment.

Abstract: Modern processor architectures sacrifice timing predictability for increased average computational throughput. Branch prediction, multi-level memory hierarchies, out-of-order execution, and data forwarding all make accurate execution time predictions impossible. As accurate timing predictions are required for task scheduling, our goal is the development of a Precision Timed (PRET) processor, along with the inter-process communication methods and operating system services, which remove many of the sources of timing indeterminacy.

Abstract: Wireless sensor-actuator networks (WSAN) are systems consisting of numerous sensing and actuation devices that interact with the environment and coordinate their activities over a wireless communication network. WSANs represent an important class of cyber-physical system (CPS) found in the national power grid and air/traffic networks. This project addresses the issue of resilience in WSANs.

Abstract: Wireless sensor-actuator networks (WSAN) consist of numerous sensing and actuation devices that share information over an ad hoc wireless communication network. WSANs can be used to manage networked systems that distribute goods and services over large spatially distributed domains. Examples of such systems include the national power grid, ground/air traffic networks, and water/gas distribution networks.

Abstract: We are developing a new computational framework and physical platform for modeling, analyzing, and designing dense networks of micro-robotic swarms. The physical platform is based on a bio-hybrid micro-robotic approach, in which bacteria serve as on- board actuators. The micro-robots are controlled through passive (e.g. chemical gradients) and active (e.g. magnetic fields) steering mechanisms. Here, we present the first step towards passive control by characterizing the chemotactic behavior of free- swimming bacteria.

Abstract: The objective of this research is to develop numerical techniques for solving partial differential equations (PDE) that govern information flow in dense wireless networks. Despite the analogy of information flow in these networks to physical phenomena such as thermodynamics and fluid mechanics, many physical and protocol imposed constraints make information flow PDEs unique and different from the observed PDEs in physical phenomena.

Abstract:  Reinforcement Learning (RL) is a paradigm for learning decision- making tasks that could enable cyber-physical systems (CPS) to learn and adapt to situations on-line. For an RL algorithm to be practical for CPS control tasks, it must learn in very few samples, while continually taking actions in real-time. In addition, the algorithm must learn effi- ciently in the face of noise, sensor/actuator delays and continuous state features. We describe TEXPLORE, a model-based RL method that ad- dresses these issues.