Abstract: The use of robots in society could be expanded by using reinforcement learning (RL) to allow robots to learn and adapt to new situations on-line. RL is a paradigm for learning sequential decision making tasks, usually formulated as a Markov Decision Process (MDP). For an RL algorithm to be practical for robotic control tasks, it must learn in very few samples, while continually taking actions in real-time. In addition, the algorithm must learn efficiently in the face of noise, sensor/actuator delays, and continuous state features.

Abstract: Securing critical networked cyber-physical systems (NCPSs) such as the power grid or transportation systems has emerged as a major national and global priority. The networked nature of such systems renders them vulnerable to a range of attacks both in cyber and physical domains as corroborated by recent threats such as the Stuxnet virus.

Abstract: The goal of this research project is to create a scalable and robust cyber-physical system (CPS) framework for the observation and control of the functional interdependencies between bridge structures (stationary physical systems) and trucks, vehicles and human inspectors (mobile physical agents). Figure 1 shows the SHM-based CPS framework that is built around demonstration highway bridges with permanent wireless bridge monitoring system installed.