Title: Optimization-based Approaches to Safety-Critical Control, with Applications to Space Systems

Date: Wednesday, July 21st, 2021

Time: 1:00 pm - 3:00 pm (EST)

Location: Virtual

Virtual access: BlueJeans

Link: https://bluejeans.com/284937699/1815

Meeting ID: 284 937 699

Committee Members:

Dr. Magnus Egerstedt (Advisor),  School of Electrical and Computer Engineering, Georgia Institute of Technology

Dr. Eric Feron (Advisor), School of Electrical and Computer Engineering, King Abdullah University of Science and Technology

Dr. Sam Coogan,  School of Electrical and Computer Engineering, Georgia Institute of Technology

Dr. Panagiotis Tsiotras, School of Aerospace Engineering, Georgia Institute of Technology 

Dr. Mark Costello, School of Aerospace Engineering, Georgia Institute of Technology 

Dr. Kerianne Hobbs, United States Air Force Research Laboratory 

Abstract:

This thesis investigates the problem of safety-critical control for complex cyber-physical systems, with an emphasis on numerical optimization and autonomy applications in the space domain. First, the concept of safety is formalized for input-constrained dynamical systems. Safety properties appear as state constraints and system safety is defined with respect these constraints using the notion of forward invariance. Next, the research moves toward identifying a general theory of Run Time Assurance (RTA). RTA relates to a control system architecture where a performance-oriented element is augmented with a safety-driven element that filters the control signal in such a way that precludes harmful decisions. Importantly, the adoption of an RTA architecture allows for research on both performance and safety fronts to be pushed in parallel, and as such the results in this research are complementary to results in performance-based research. The latter part of the thesis consists of application-specific research for various problems on space systems, including autonomous rendezvous and docking subject to collision avoidance constraints, attitude control subject to line-of-sight constraints, and collision-inclusive planning for free-flying intravehicular robots.