Hey there 👋,
this is a simple application of the Q-learning algorithm in a "Find the exit" simulation game.
This project is part of my bachelor's thesis where I explored the state-of-the-art of AI algorithms in video games.
The modeled entities are:
- The escapee (as the Red dot)
- The escapee skills (a set K of movement actions)
- The maze (an NxM matrix of walls, trap-pits, neutral tiles and finally, the exit)
The final goal is to find the shortest path to the exit using only the available movement skills (for example, you can only move like the knight in chess).
There are some hard-to-watch violations of the SOLID principles, but I choose to freeze the project as it was, as a memento of a long-gone past 😢