The EUROCONTROL CAFÉ platform is a set of tools for running European validation exercises on Airborne Collision Avoidance Systems (ACAS) and Detect And Avoid (DAA) systems. CAFÉ includes tools for building statistical encounter models to generate significantly large sets of close encounters and a Monte Carlo fast-time simulator CAVEAT (Collision Avoidance Validation and Evaluation Tool) for simulating ACAS and DAA along those encounters. Four statistical encounter models have been built so far:
CAFÉ Revised Encounter Model for Europe (CRÈME) based on 12 million flight hours of radar data from nine European countries collected 2015-18 for generating representative encounters.
Multi-Aircraft Encounters by STochastic RecOnstruction (MAESTRO) is a multi-aircraft encounter model that combines data from real pairwise encounters to generate encounters for testing performance with several intruders.
SA01 generates Überlingen-like encounters with various pairwise vertical chase scenarios.
Stress model generates extreme encounters designed to test robustness.
CAFÉ allows quantitative performance assessment of risk and operational acceptance metrics with statistical significance. CRÈME and CAVEAT were used for the European validation of ACAS Xa to demonstrate with 95% confidence that alerts would be reduced by 60% and safety improved by 20% compared with the current system TCAS II (Traffic alert and Collision Avoidance System). The variant CRÈME RPAS (Remotely Piloted Aircraft Systems) has been used to generate large encounter sets for SESAR projects ERICA (Enable RPAS Insertion in Controlled Airspace, airspace classes A-C) and URCLEARED Remain Well Clear for integration of RPAS in airspace classes D-G.