Data-Driven Safety-Critical Control: Synthesizing Control Barrier Functions With Koopman Operators

Control barrier functions (CBFs) are a powerful tool to guarantee safety of autonomous systems, yet they rely on the computation control invariant sets, which is notoriously difficult. A backup strategy employs an implicit set computed by forward integrating system dynamics. However, this integration prohibitively expensive for high dimensional and inaccurate in presence unmodelled We propose learn discrete-time Koopman operators closed-loop dynamics under strategy. This approach replaces simple matrix multiplication, can mostly be offline. also derive error bound unmodeled order robustify CBF controller. Our extends multi-agent we demonstrate method collision avoidance wheeled robots quadrotors.