Learning for Layered Safety-Critical Control with Predictive Control Barrier Functions
This addresses safety-critical control for complex systems like robots, but it is incremental as it builds on existing CBF methods.
The paper tackled the problem of safety violations due to gaps between reduced-order and full-order models in safety-critical control by introducing predictive control barrier functions (CBFs) that use full-order model rollouts to add a predictive robustness term. The result demonstrated safe behavior with minimal conservatism in simulation and was experimentally validated on a 3D hopping robot.
Safety filters leveraging control barrier functions (CBFs) are highly effective for enforcing safe behavior on complex systems. It is often easier to synthesize CBFs for a Reduced order Model (RoM), and track the resulting safe behavior on the Full order Model (FoM) -- yet gaps between the RoM and FoM can result in safety violations. This paper introduces \emph{predictive CBFs} to address this gap by leveraging rollouts of the FoM to define a predictive robustness term added to the RoM CBF condition. Theoretically, we prove that this guarantees safety in a layered control implementation. Practically, we learn the predictive robustness term through massive parallel simulation with domain randomization. We demonstrate in simulation that this yields safe FoM behavior with minimal conservatism, and experimentally realize predictive CBFs on a 3D hopping robot.