Reinforcement Learning for Aviation Training

This project seeks to explore the use of a class of artificial intelligence algorithms called reinforcement learning for the purpose of aiding the training of new pilots. In the process, we seek to “teach” an algorithm how to fly an aircraft by exposing the AI pilot to a virtual environment and providing it with flight data and a goal. Alternatively, the algorithm could learn by observing human pilots. If this approach is successful, it could lead to better autopilot systems as well as teaching aids for new pilots by simulating their response to different flying environments and helping instructors select the most appropriate training exercises.

Faculty Supervisor:

Yoshua Bengio

Student:

Sahar Bahrami

Partner:

Paladin AI Inc

Discipline:

Computer science

Sector:

Aerospace and defense

University:

Université de Montréal