Error Detection within a Specific Time Horizon and Application to Air Traffic Management

We propose a novel concept of observability for hybrid systems, critical observability, for estimating and mitigating the probability of errors in Air Traffic Management. Critical observability refers to states that are characterized by the existence of an evolution that may yield catastrophic situations. Detecting these states is an important element of a safety critical environment and we design a critical observer that we apply to the runway crossing problem. We also present an extension of this theory to a class of stochastic hybrid systems and apply it to a clearance changing the flight plan.