Certification of Dynamical Systems using End-to-End Distributionally-Robust Uncertainty Quantification

We present OptimalUncertaintyQuantification.jl: A Julia tool for end-to-end distributionally robust uncertainty quantification of static and dynamical systems models. It provides an ergonomic symbolic interface for modeling the system as well as principled inclusion of data and knowledge about uncertainities. Furthermore, it automatically transforms the certification problem into a computationally tractable optimization problem solvable by integrated solvers. We demonstrate this tool on the benchmark Moore-Greitzer Ordinary Differential Equation model and provide an iterative workflow for model-based systems designers to meet safety certification targets.