Future airliners with artificial intelligence will bring a smooth in-flight experience

Airplane turbulence can be eliminated with FALCON, a new artificial intelligence system that helps vehicles learn how to control themselves. Scientists have developed a technique that can reduce the effects of turbulence on dynamic structures and vehicles, with a special focus on unmanned aerial vehicles (UAVs).

Turbulence is the name we give to changes in air pressure that cause the plane to shake; More turbulence occurs when an aircraft is experiencing changes in air pressure mid-flight. Unlike flying animals, airliners have evolved a natural ability to stay calm during flight.

Research published in the journal NPJ Robotics showed how scientists could develop a control technique for passenger aircraft. This technique required the use of an artificial intelligence (AI) system called FALCON to automatically adjust the flight to compensate for turbulence.

FALCON is trained with reinforcement learning to understand the fundamentals that cause turbulence to adapt to any situation. The researchers found that digitally representing wind conditions as periodic waves is an effective tool for turbulence modeling because wind tides and their effects naturally follow a wave pattern.

Scientists tested the AI ​​in a wind tunnel at Caltech by installing it along with pressure sensors and control surfaces on a drone. The AI ​​used these tools to sense pressure changes and adjust its pitch and yaw as needed to maintain stability. A moving cylinder was also placed upstream in the tunnel to generate random fluctuations in the turbulence.

It was found that after 9 minutes of learning, where FALCON continuously tries to adapt to the changing turbulence and feedback the results, the AI ​​can be stable.

Caltech’s wind tunnel tests showed that FALCON can learn in minutes, indicating scalability for larger aircraft. However, real-world challenges remain, especially in quickly adapting to diverse and unpredictable conditions and validating performance in different UAV configurations and wind environments.

This research has the potential to lead to smoother flight for drones and commercial aircraft. Researchers have also proposed the possibility of sharing environmental data between aircraft in order to warn of disturbances. However, given the cybersecurity concerns surrounding aircraft control systems, this requires a robust security protocol that must be thoroughly vetted and tested in advance.