Abstract
Safety has been one of the key objectives of controlling dynamical systems. Recently, safety for data-driven control of dynamical systems has raised much attention. This chapter introduces how to use data-driven approach for safe learning and control. First, we introduce the forward invariance, which is the key component for ensuring safety. Then we introduce how to design a constrained optimal control problem (OCP) to solve the safe controller. We show two types of definitions of forward invariance, control barrier function and reachability, to design the safe control problem as well as data-driven methods to learn the forward invariant sets. Then we introduce learning-based methods to learn the safe controller. Finally, we discuss the dilemma of learning-based safety certificates and safe controller, and propose a joint learning algorithm for safety certificates and safe control policy with theoretical guarantees.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Reference Module in Materials Science and Materials Engineering
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
