Learning

The ATM system is a complex system composed of human actors, computer tools, various operational procedures. It is daily subject to random external factors such as weather.

The information covered in this system are usually shrouded in uncertainty. Some are sometimes difficult to quantify, such as the workload of the air traffic controller.

In a number of cases, it is possible to learn this information from observable variables, using machine learning techniques (Machine Learning). This approach can be an alternative to the statistical approach (see the Probability et Statistics page).

Some examples

Machine Learning is still relatively little used in ATM. The number of potential applications is nevertheless important.

Here are some examples of work done by MAIAA:

This work is the result of close collaboration with teams APO and Adria at the Institute for Computer Science of Toulouse (IRIT).