How to learn control policies for complex dynamic systems from data with safety and security guarantees?
Science / Mathematics
Control systems are fundamental to cyber-physical systems (CPS), engineered systems that integrate computational and physical components. These systems rely heavily on communication and computing technologies. As CPS become more commonplace, society’s dependence on them grows, making security and reliability critical. Fortunately, these technologies offer benefits by streamlining processes, increasing flexibility, and providing a wealth of data. This data can be used to address the inherent complexity of these integrated systems. This project proposes to use this wealth of data to design reliable control systems for CPS and to provide certifications that ensure their safe and reliable operation, even in the face of cyber threats arising from their underlying technologies.
