Search
The main task of the homework is to implement an advanced player policy for pursuit-evasion game.
.py
player.py
eval.py
game.py
pacman.policy
mazes/pacman.policy
self.timeout
def monte_carlo_policy(self, grid_map, evaders, pursuers): self.next_robots = self.robots[:] #for each robot plan actions for idx in range(0, len(self.robots)): clk = time.time() while (time.time() - clk) < self.timeout: #monte-carlo tree search #make decision on where to go based on tree search self.next_robots[idx] = pos_selected
The assignment will be evaluated within the provided game simulation framework.
pacman.game
mazes/pacman.game
EVADER VALUE_ITERATION ro 1 1 PURSUER VALUE_ITERATION bo 3 3 17 19
EVADER MONTE_CARLO ro 1 1 PURSUER MONTE_CARLO bo 3 3 17 19
grid4x4