redbluedoors.py 2.0 KB

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465
  1. from gym_minigrid.minigrid import Door, Grid, MiniGridEnv
  2. class RedBlueDoorEnv(MiniGridEnv):
  3. """
  4. Single room with red and blue doors on opposite sides.
  5. The red door must be opened before the blue door to
  6. obtain a reward.
  7. """
  8. def __init__(self, size=8, **kwargs):
  9. self.size = size
  10. super().__init__(
  11. width=2 * size, height=size, max_steps=20 * size * size, **kwargs
  12. )
  13. def _gen_grid(self, width, height):
  14. # Create an empty grid
  15. self.grid = Grid(width, height)
  16. # Generate the grid walls
  17. self.grid.wall_rect(0, 0, 2 * self.size, self.size)
  18. self.grid.wall_rect(self.size // 2, 0, self.size, self.size)
  19. # Place the agent in the top-left corner
  20. self.place_agent(top=(self.size // 2, 0), size=(self.size, self.size))
  21. # Add a red door at a random position in the left wall
  22. pos = self._rand_int(1, self.size - 1)
  23. self.red_door = Door("red")
  24. self.grid.set(self.size // 2, pos, self.red_door)
  25. # Add a blue door at a random position in the right wall
  26. pos = self._rand_int(1, self.size - 1)
  27. self.blue_door = Door("blue")
  28. self.grid.set(self.size // 2 + self.size - 1, pos, self.blue_door)
  29. # Generate the mission string
  30. self.mission = "open the red door then the blue door"
  31. def step(self, action):
  32. red_door_opened_before = self.red_door.is_open
  33. blue_door_opened_before = self.blue_door.is_open
  34. obs, reward, done, info = MiniGridEnv.step(self, action)
  35. red_door_opened_after = self.red_door.is_open
  36. blue_door_opened_after = self.blue_door.is_open
  37. if blue_door_opened_after:
  38. if red_door_opened_before:
  39. reward = self._reward()
  40. done = True
  41. else:
  42. reward = 0
  43. done = True
  44. elif red_door_opened_after:
  45. if blue_door_opened_before:
  46. reward = 0
  47. done = True
  48. return obs, reward, done, info