If you are an ANSP dealing with high controller workload and complexity — this project developed an AI Assistant Application that tracks visual attention to provide human-centric support. This reduces the mental load on controllers while keeping them active in the loop.
AI Assistant for Air Traffic Control that Understands Human Intent and Attention
Imagine a co-pilot that doesn't just follow orders but actually notices where you are looking and what you are trying to do. Instead of the human having to explain everything to the computer, the computer watches the human's eyes and actions to guess their goals. This helps the AI step in exactly when needed to reduce stress and prevent mistakes during busy flights.
What needed solving
Air traffic controllers face high mental loads in complex scenarios, and current AI tools often act as 'black boxes' that don't understand what the human is trying to achieve. This leads to a dangerous gap where humans either over-rely on automation or struggle to take over when the system fails.
What was built
An AI Assistant Application that uses visual attention tracking and interaction analysis to recognize controller intent and provide adaptable support.
Who needs this
Who can put this to work
If you are a software developer dealing with the risk of automation bias where humans lose their skills — this project developed a system that recognizes human intent to provide adaptable support. This ensures controllers maintain their expertise rather than becoming passive supervisors.
If you are a command center dealing with high-stress environments and complex data streams — this project developed artificial situational awareness that detects when a human loses focus. This allows the system to bring the operator back into the loop and prevent critical errors.
Quick answers
What is the cost or price of implementing this AI assistant?
Based on available project data, there is no specific commercial pricing provided; however, the EU has contributed EUR 1,940,625 to the research and development phase.
Is this system ready for industrial scale deployment?
The project is currently in the development and testing phase, with initial work based on theoretical research and experimental scenarios. It is not yet at full industrial scale.
How is the IP and licensing handled for the AI modules?
Based on available project data, specific licensing terms are not mentioned, but the project involves a consortium of 9 partners across 9 countries.
How does this integrate with existing air traffic control workstations?
The system integrates by tracking visual attention and analyzing how the controller interacts with their work station to determine intent.
What is the timeline for the completion of the system?
The project is scheduled to run from 2024-06-01 to 2026-11-30.
Who built it
The consortium is well-balanced for a research-to-industry transition, consisting of 9 partners from 9 different countries. With 3 industry partners (33% ratio) and 4 universities, the project blends academic rigor with practical application. The presence of SMEs and large entities suggests a focus on both agile development and large-scale aviation standards.
Contact SVEUCILISTE U ZAGREBU FAKULTET PROMETNIH ZNANOSTI in Croatia
Talk to the team behind this work.
Contact us to find partnership opportunities with the AWARE consortium.