DIALOG aims to is to enable efficient collaboration between air traffic controllers (ATCO) and AI, by delivering novel teamwork solutions for human-AI teams. DIALOG will enable inferring ATCOs’ intent and goals through the use of task models developed by applying speech recognition and understanding of pilot-controller exchanges on the radio frequency. Furthermore, DIALOG will develop ML-based methods for unobtrusive real-time assessment of ATCOs workload and attention based on voice, physiological markers, behavioural data and their combinations.
A digital assistant, called Teamwork assistant will enable effective and trustworthy human-AI teaming. It will determine how to best support ATCOs by assigning tasks to AI assistants that might either help ATCOs to perform a task or do it autonomously. The Teamwork assistant’s actions will depend on the ATCOs’ current state, attention allocation, traffic situation, and general context. Models, methods and prototypes developed in the project will be validated in several use cases.
SINTEF is the project coordinator and a research partner. SINTEF Human-Computer Interaction Group brings in its expertise in design and evaluation of systems for complex work situations, as well as expertise in AI-enabled decision support.