Imagine trying to get a driver's license for a self-driving car, but there are no rules yet on how to test if it's actually safe. This work creates the rulebook and a toolkit for building AI that helps air traffic controllers manage planes without getting overwhelmed. It ensures that as machines take over more tasks, humans still stay in control and the system remains legal and safe.
Aviation companies cannot easily deploy AI because there are no clear, legal rules for certifying that these systems are safe and reliable. This creates a bottleneck for innovation in autonomous flight and air traffic control.
A certification method for aviation authorities and a design toolkit for manufacturers to build AI-powered ATM systems.
If you are a software developer dealing with the difficulty of getting AI-powered flight systems approved by regulators — this project developed a set of design guidelines and a toolkit that streamlines the development process to ensure legal compliance.
If you are an ANSP dealing with high controller stress and workload — this project developed a unified certification method that allows you to safely integrate advanced automation into your ground systems.
If you are a drone traffic operator dealing with the lack of standards for autonomous airspace — this project developed a certification approach specifically targeting U-Space and urban air mobility to accelerate innovation.
Based on available project data, no pricing or licensing costs are mentioned; the project is funded by an EU contribution of EUR 998,900.
Yes, the project targets industrial manufacturers and national aviation authorities to ensure the guidelines are applicable to real-world airborne and ground systems.
Based on available project data, the IP ownership is not specified, though it is developed by a consortium led by DEEP BLUE SRL.
It aims to fill gaps in existing regulations by creating a method for the approval and certification of AI and machine learning in ATM environments.
The project period runs from 2023-09-01 to 2026-02-28, suggesting results will be finalized by early 2026.
The consortium is lean and highly specialized, consisting of 6 partners across 3 countries (DE, IT, NL). It features a strong industrial lean with a 33% industry ratio, including 2 industry partners and 1 SME (the coordinator, DEEP BLUE SRL), balanced by 3 research entities and 1 university. This structure suggests a direct pipeline from academic research to industrial application.
Contact DEEP BLUE SRL in Italy for details on the design toolkit.
Contact us to get the design guidelines for AI aviation certification.
