redbluedoors.py 2.3 KB

1234567891011121314151617181920212223242526272829303132333435363738394041424344454647484950515253545556575859606162636465666768697071
  1. from gym_minigrid.minigrid import Door, Grid, MiniGridEnv, MissionSpace
  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. mission_space = MissionSpace(
  11. mission_func=lambda: "open the red door then the blue door"
  12. )
  13. super().__init__(
  14. mission_space=mission_space,
  15. width=2 * size,
  16. height=size,
  17. max_steps=20 * size * size,
  18. **kwargs
  19. )
  20. def _gen_grid(self, width, height):
  21. # Create an empty grid
  22. self.grid = Grid(width, height)
  23. # Generate the grid walls
  24. self.grid.wall_rect(0, 0, 2 * self.size, self.size)
  25. self.grid.wall_rect(self.size // 2, 0, self.size, self.size)
  26. # Place the agent in the top-left corner
  27. self.place_agent(top=(self.size // 2, 0), size=(self.size, self.size))
  28. # Add a red door at a random position in the left wall
  29. pos = self._rand_int(1, self.size - 1)
  30. self.red_door = Door("red")
  31. self.grid.set(self.size // 2, pos, self.red_door)
  32. # Add a blue door at a random position in the right wall
  33. pos = self._rand_int(1, self.size - 1)
  34. self.blue_door = Door("blue")
  35. self.grid.set(self.size // 2 + self.size - 1, pos, self.blue_door)
  36. # Generate the mission string
  37. self.mission = "open the red door then the blue door"
  38. def step(self, action):
  39. red_door_opened_before = self.red_door.is_open
  40. blue_door_opened_before = self.blue_door.is_open
  41. obs, reward, terminated, truncated, info = super().step(action)
  42. red_door_opened_after = self.red_door.is_open
  43. blue_door_opened_after = self.blue_door.is_open
  44. if blue_door_opened_after:
  45. if red_door_opened_before:
  46. reward = self._reward()
  47. terminated = True
  48. else:
  49. reward = 0
  50. terminated = True
  51. elif red_door_opened_after:
  52. if blue_door_opened_before:
  53. reward = 0
  54. terminated = True
  55. return obs, reward, terminated, truncated, info