La Sapienza - Università di Roma

TRUSTY is a research project aimed to exploit the power of Explainable Artificial Intelligence (XAI). Explainable Artificial Intelligence (XAI) is a field within Artificial Intelligence (AI) that focuses on developing models and systems capable not only of making decisions but also of explaining their functionality and the reasoning behind their choices.

The project’s goal is to develop new XAI-based solutions for remote digital towers (RDT) focused on specific tasks of tower controllers, such as monitoring taxiways and runways for takeoff and landing.

Unlike physical towers, where operators have direct visual access to air traffic, remote digital towers provide such information through video transmissions. Integrating XAI into these towers aims to ensure that safety and efficiency remain equivalent to those of traditional towers while making automated decision-making processes transparent and interpretable.

The XAI developed within TRUSTY will allow operators to understand not only the system’s autonomous decisions but also the logic and reasoning behind those decisions. This is crucial for maintaining a high level of trust and control in automation systems, especially in high-risk contexts where interaction with AI must be completely clear and understandable.

The project will therefore develop a reliable XAI system for RDTs that meets not only requirements for adaptability, accuracy, and accountability but also transparency, ensuring safe and interpretable decision support.

The Sapienza Lab team will participate in tests conducted in real contexts to validate the safety and reliability performance of the system, addressing the needs of primary user groups even within simulated RDT environments.

Participants:

Malardalens Universitet,

Deep Blue,

Ecole Nationale De L' Aviation Civile (ENAC),

Universita degli Studi Di Roma La Sapienza Laboratorio Neuroscienze Industriali F. Babiloni