Maltese researchers are working on a new artificial intelligence and machine learning-based system that will learn to fly like human pilots.

The system is intended to carry out high-level tasks, such as interpretation of weather information and the calculation of weather-avoidance manoeuvres, which are currently not automated on planes.

“The proposed system will act as an 'artificial pilot' – supporting human pilots in challenging situations and, consequently, improving flight safety,” the University of Malta said in a statement.

It would  also help pilots recover from loss-of-control situations which go beyond the capabilities of current automation, such as aerodynamic stalls and spins.

On-board automation systems already existed, but pilots were left to handle the aircraft on their own should a critical sensor fail or should the aircraft enter an abnormal state. Most recently, a Boeing 737 Max crashed minutes into a flight, killing 157 people on board.

The system would seek to improve aircraft automation, the statement said, adding that pilots had to combine information from multiple sources and needed to reroute the aircraft should severely bad weather, such as a thunderstorm, be detected.

“This requires the pilots to interpret the weather situation and to figure out the best way to avoid bad weather along their route,” the statement said.

The €195,000 project is funded by the Malta Council for Science and Technology.