- Building a randomized maze generator to create a survival-game-liked environment
- Building a knowledge base for the logical agent
- Adding the agent's ability to prioritize and infer actions from its perception using pyPL which is a propositional logic interface/interpreter that I wrote from scratch https://github.com/thyton/pyPL
- Adding a modified A* algorithm to help the agent travel through the environment optimally
To run the demo agent in a 4 by 4 maze:
- Under the repo root directory, run
git clone https://github.com/thyton/pyPL
- Run
python game.py
(make sure you use Python 3.6+)