The AWARE project develops a solution designed to enhance human-machine collaboration in the air traffic control environment. This solution focuses on establishing shared situational awareness and sets the groundwork for further ML-based automation in air traffic management. For this the following elements are developed: (1) an enhanced artificial situational awareness system is developed by implementing methods for assessment of ATCO’s intent and goals, and (2) a method for the adaptable selection and implementation of actions that support ATCOs, utilizing both ML and non-ML tools.

This requires the capability to track human visual attention in combination with other inputs on the human-machine interface and to give them semantics by putting them in context of the current traffic situation. The project will research methods of identifying loss of situational awareness (ranging from inadequate selectivity to degraded situation awareness to complete out-of-the-loop state) and exploring options for bringing the human back into the loop. Furthermore the project will define specifications for interoperability with other systems and roles in ATM (e.g. ATSEP, FMP) in order to explore benefits of artificial situational awareness system beyond ATC tactical operations, including benefits to the certification process, cybersecurity benefits and novel threats, and possible improvements on pre-tactical operations.

The goal of the project is to enable human-machine collaboration by using an artificial situational awareness system, which enables AI to anticipate and respond to human needs by understanding human intent and goals. While humans are extensively trained to understand the capabilities, limitations, and functionality of the machines they are using, further improvements in human-machine collaboration are currently hindered by a lack of awareness of human intent on the side of machines.

The project will develop and test an AI Assistant Application that provides adaptable human-centric support to enhance air traffic controllers' (ATCOs) performance and reduce ATCOs' workload despite high task complexity. This will be achieved by developing a human-machine collaboration environment that relies on the recognition of ATCO intent, ATCO situational awareness (compared to machine situational awareness), and ATCO mental load. ATCO intent will be analyzed by tracking their attention and human-machine interactions and comparing them to the tasks that need solving, as assessed by the artificial situational awareness system.

Adaptable support will then be provided either by assisting with the task they are currently focused on or by autonomously solving an unrelated task. This will allow ATCOs to maintain their skills and expertise while preventing a shift towards supervisory control, which has been demonstrated to undermine human capability to take over in situations with degraded automation.

A key goal of the adaptable and human-aware system is to keep ATCOs in an active role, helping them train their skills and expertise on the job while selectively using higher levels of automation to augment capacity. ATCOs are supported in their tasks rather than replaced by automation. It is expected that ATCOs will be able to handle high-complexity scenarios more effectively when assisted by an attention-aware support system. ATCO workload is expected to decrease with the use of support functions.