Is a project for an open-source reinforcement learning toolkit for autonomous fixed wing aircraft simulation developed by David Richter and Ricardo Calix. For more information visit the following link:


A sister implementation is Deep Q AP 2.0 which can be accessed with the link below.

Deep Q AP 2.0

Deep Q AP 2.0 (Deep Q Auto Pilot) is an implementation of DQN learning for automated flight using x-plane 11 and xpc. This version (2.0) provides the basic pipeline for the environment and 1 module for aircraft in flight stabilization. Basic DQN (Deep Q Network) learning functionality is also provided.


This is a video example of QPlane: