Title: ARTIAP (Artificial Intelligence-based Additional Pilot)

Duration: 2024-2026

Funding scheme: MCST FUSION Technology Development Programme (R&I-2021-014-T)

Funds: EUR 294,865

Principal investigator: Dr Ing. Jason Gauci

Automation on current transport category aircraft has evolved to the point where the pilot’s role is becoming increasingly supervisory in nature. However, this has led to pilots becoming less aware of how the automation is behaving, posing a risk to the continued safety of flight. In addition, the industry is moving towards a reduction in flight crew through the introduction of Reduced Crew Operations (RCOs) in the near-term and Single Pilot Operations (SPOs) in the long-term. This poses challenges, particularly in high workload conditions, and the support currently provided by the second pilot needs to be replaced by other means.

ArtiAP will develop technology - based on Artificial Intelligence (AI) and Machine Learning (ML) - to provide a replacement of the support traditionally provided by the second pilot and to also provide support to the current two-person crew in large transport aircraft. The ArtiAP technology will monitor complex operations in flight - such as a late runway change or an unstable approach - and aid the pilots in their situational awareness and decision-making by providing visual and/or aural alerts and recommendations. In addition, the crew will be able to interact with the AI/ML using voice commands and touchscreen gestures. The key advantages of the proposed technology are that it: keeps the flight crew in the decision loop; is compatible with (and augments) existing cockpit automation (autopilot, etc.); and acts as an additional (artificial) pilot in the cockpit.

The project will first identify and analyse specific use cases where pilots would benefit the most from the application of AI/ML. Then, it will design, develop, train and test ML models for each of these use cases, using a combination of supervised, unsupervised and reinforcement learning techniques. In addition, the project will design and develop a Human-Machine Interface (HMI) to enable multi-modal interaction between the pilots and the AI. Finally, the ArtiAP technology will be demonstrated and evaluated in a realistic flight simulation environment, with the participation of airline pilots, in order to assess its performance and end-user acceptability.