First Release
Fulfills the original goal of the project, which was to use reinforcement learning to get a Space X style rocket to land itself. Contains training and demo scenes, as well as a trained model for the lander.
Features
- Three scenarios: missile, hover, and lander
- Training and demo scenes for all three
- Trained neural network models for all three
- Using a gamepad to manually control the rocket, or switch to AI control
- Multiple cameras in the lander demo scene, with the ability to switch between them