Travail Pratique

TP1: Candy Grab Game: check who is the best AI player? can you suggest a good one?

source code: https://github.com/hmitiche/candygrabgame

TP 2: Vacuum Cleaner Robot
Watch the following simulation of two vacuum cleaner agents in a 4 rooms house (wrote in Python, open AI Gymnasium and Pygame):
0- repeat this experiment , download the code (link below) and run in terminal:

python testme.py
1 - the agents cleaned almost the same number of rooms, why the scores are different?
1- describe the behaviour of each agent?
2- write the pseudo-code of each agent (agent function)? (reverse-engineering)
3- write the pseudo-code of a  rational agent (best-you-can) for the case wherethe  robot knows ahead
that dirt won't comeback?  what is the type of this agent (reflex simple, reflex random or reflex with model)?

PS: the code of the agent can be found in the repo in: policy/greedy.py, from line 307 (or where
there is :"def agent_4rooms_v1(self, location, dirty): ..."

simulator code: https://github.com/hmitiche/vacuumCleaner

Simulator code in Python:  https://github.com/hmitiche/vacuumCleaner