TRUSTY, a new SESAR 3 JU project, recently kicked-off aimed at implementing artificial intelligence (AI) in remote digital towers thanks to a trusted intelligent system.

Meeting on 6 September in Västerås, at the Mälardalen University - coordinator of the research project – the project's success counts on the collaboration of a multidisciplinary consortium comprising four partners from three countries: Sweden, Italy and France. The consortium includes three academic institutions, and a research and consultancy company. The partners are complementary to each other and bring a combination of academic, technical, human factor and operational skills and expertise to the project.

Remote digital towers represent the latest innovation in aviation, offering remote traffic flow and capacity management for airports. While conventional control tower host operators have direct visual oversight of runways and taxiways, digital towers harnesses video transmission to provide the same vital information. This advancement enables the provision of airport air traffic services (ATS) from virtually anywhere, promising significant enhancements in operational efficiency and safety by augmenting controller situational awareness.

In today's era, artificial intelligence and machine learning are offering automated and faster solutions in many industries, bringing the industry to a more advanced stage. However, in a safety-critical industry like air traffic management, the adoption of these technologies calls for transparent and trustworthy applications. “TRUSTY's overarching objective is to elevate the transparency and comprehensibility of AI-powered decisions within the context of RDT, thereby establishing trustworthiness and reliability,” says Prof. Mobyen Uddin Ahmed, coordinator of TRUSTY and expert in explainable artificial intelligence.

TRUSTY aims at increasing the use of automated technology, support air traffic controllers in repetitive and routine tasks. The ambition is to enable controllers to focus on safety-critical and complex tasks. In this regard, having algorithms which are transparent, reliable and explainable can make a big difference in the implementation of them.

TRUSTY's success relies on state-of-the-art developments in interactive data visualisation and user-centric explanation, coupled with recent technological advancements in AI accuracy, robustness, interpretability, fairness, and accountability. The project incorporates information visualization techniques such as visual analytics, data-driven storytelling, and immersive analytics in human-machine interactions (HMI). As a result, TRUSTY stands at the intersection of trustworthy AI, multi-model machine learning, active learning, and user experience (UX) for human and AI model interaction.

“The potential results will be a proof-of-concept of a trustworthy intelligent system for RDTs which will lead to higher automation to ensure safe and reliable decision support” says Prof. Mobyen.

Co-funded within the framework of Horizon Europe and SESAR 3 JU. TRUSTY is a collaborative project involving the following organisations:

1. Mälardalen University (MDU, SE)

2. Ecole Nationale de l'Aviation Civile (ENAC, FR)

3. Deep Blue Srl (DBL, IT)

4. Università di Roma "La Sapienza" (UNIROMA1, IT)