obstructedmaze.py 7.3 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257
  1. from gym_minigrid.minigrid import COLOR_NAMES, DIR_TO_VEC, Ball, Box, Key
  2. from gym_minigrid.register import register
  3. from gym_minigrid.roomgrid import RoomGrid
  4. class ObstructedMazeEnv(RoomGrid):
  5. """
  6. A blue ball is hidden in the maze. Doors may be locked,
  7. doors may be obstructed by a ball and keys may be hidden in boxes.
  8. """
  9. def __init__(self, num_rows, num_cols, num_rooms_visited, **kwargs):
  10. room_size = 6
  11. max_steps = 4 * num_rooms_visited * room_size**2
  12. super().__init__(
  13. room_size=room_size,
  14. num_rows=num_rows,
  15. num_cols=num_cols,
  16. max_steps=max_steps,
  17. **kwargs
  18. )
  19. def _gen_grid(self, width, height):
  20. super()._gen_grid(width, height)
  21. # Define all possible colors for doors
  22. self.door_colors = self._rand_subset(COLOR_NAMES, len(COLOR_NAMES))
  23. # Define the color of the ball to pick up
  24. self.ball_to_find_color = COLOR_NAMES[0]
  25. # Define the color of the balls that obstruct doors
  26. self.blocking_ball_color = COLOR_NAMES[1]
  27. # Define the color of boxes in which keys are hidden
  28. self.box_color = COLOR_NAMES[2]
  29. self.mission = "pick up the %s ball" % self.ball_to_find_color
  30. def step(self, action):
  31. obs, reward, done, info = super().step(action)
  32. if action == self.actions.pickup:
  33. if self.carrying and self.carrying == self.obj:
  34. reward = self._reward()
  35. done = True
  36. return obs, reward, done, info
  37. def add_door(
  38. self,
  39. i,
  40. j,
  41. door_idx=0,
  42. color=None,
  43. locked=False,
  44. key_in_box=False,
  45. blocked=False,
  46. ):
  47. """
  48. Add a door. If the door must be locked, it also adds the key.
  49. If the key must be hidden, it is put in a box. If the door must
  50. be obstructed, it adds a ball in front of the door.
  51. """
  52. door, door_pos = super().add_door(i, j, door_idx, color, locked=locked)
  53. if blocked:
  54. vec = DIR_TO_VEC[door_idx]
  55. blocking_ball = Ball(self.blocking_ball_color) if blocked else None
  56. self.grid.set(door_pos[0] - vec[0], door_pos[1] - vec[1], blocking_ball)
  57. if locked:
  58. obj = Key(door.color)
  59. if key_in_box:
  60. box = Box(self.box_color)
  61. box.contains = obj
  62. obj = box
  63. self.place_in_room(i, j, obj)
  64. return door, door_pos
  65. class ObstructedMaze_1Dlhb(ObstructedMazeEnv):
  66. """
  67. A blue ball is hidden in a 2x1 maze. A locked door separates
  68. rooms. Doors are obstructed by a ball and keys are hidden in boxes.
  69. """
  70. def __init__(self, key_in_box=True, blocked=True, **kwargs):
  71. self.key_in_box = key_in_box
  72. self.blocked = blocked
  73. super().__init__(num_rows=1, num_cols=2, num_rooms_visited=2, **kwargs)
  74. def _gen_grid(self, width, height):
  75. super()._gen_grid(width, height)
  76. self.add_door(
  77. 0,
  78. 0,
  79. door_idx=0,
  80. color=self.door_colors[0],
  81. locked=True,
  82. key_in_box=self.key_in_box,
  83. blocked=self.blocked,
  84. )
  85. self.obj, _ = self.add_object(1, 0, "ball", color=self.ball_to_find_color)
  86. self.place_agent(0, 0)
  87. class ObstructedMaze_Full(ObstructedMazeEnv):
  88. """
  89. A blue ball is hidden in one of the 4 corners of a 3x3 maze. Doors
  90. are locked, doors are obstructed by a ball and keys are hidden in
  91. boxes.
  92. """
  93. def __init__(
  94. self,
  95. agent_room=(1, 1),
  96. key_in_box=True,
  97. blocked=True,
  98. num_quarters=4,
  99. num_rooms_visited=25,
  100. **kwargs
  101. ):
  102. self.agent_room = agent_room
  103. self.key_in_box = key_in_box
  104. self.blocked = blocked
  105. self.num_quarters = num_quarters
  106. super().__init__(
  107. num_rows=3, num_cols=3, num_rooms_visited=num_rooms_visited, **kwargs
  108. )
  109. def _gen_grid(self, width, height):
  110. super()._gen_grid(width, height)
  111. middle_room = (1, 1)
  112. # Define positions of "side rooms" i.e. rooms that are neither
  113. # corners nor the center.
  114. side_rooms = [(2, 1), (1, 2), (0, 1), (1, 0)][: self.num_quarters]
  115. for i in range(len(side_rooms)):
  116. side_room = side_rooms[i]
  117. # Add a door between the center room and the side room
  118. self.add_door(
  119. *middle_room, door_idx=i, color=self.door_colors[i], locked=False
  120. )
  121. for k in [-1, 1]:
  122. # Add a door to each side of the side room
  123. self.add_door(
  124. *side_room,
  125. locked=True,
  126. door_idx=(i + k) % 4,
  127. color=self.door_colors[(i + k) % len(self.door_colors)],
  128. key_in_box=self.key_in_box,
  129. blocked=self.blocked
  130. )
  131. corners = [(2, 0), (2, 2), (0, 2), (0, 0)][: self.num_quarters]
  132. ball_room = self._rand_elem(corners)
  133. self.obj, _ = self.add_object(*ball_room, "ball", color=self.ball_to_find_color)
  134. self.place_agent(*self.agent_room)
  135. class ObstructedMaze_2Dl(ObstructedMaze_Full):
  136. def __init__(self, **kwargs):
  137. super().__init__((2, 1), False, False, 1, 4, **kwargs)
  138. class ObstructedMaze_2Dlh(ObstructedMaze_Full):
  139. def __init__(self, **kwargs):
  140. super().__init__((2, 1), True, False, 1, 4, **kwargs)
  141. class ObstructedMaze_2Dlhb(ObstructedMaze_Full):
  142. def __init__(self, **kwargs):
  143. super().__init__((2, 1), True, True, 1, 4, **kwargs)
  144. register(
  145. id="MiniGrid-ObstructedMaze-1Dl-v0",
  146. entry_point="gym_minigrid.envs.obstructedmaze:ObstructedMaze_1Dlhb",
  147. key_in_box=False,
  148. blocked=False,
  149. )
  150. register(
  151. id="MiniGrid-ObstructedMaze-1Dlh-v0",
  152. entry_point="gym_minigrid.envs.obstructedmaze:ObstructedMaze_1Dlhb",
  153. key_in_box=True,
  154. blocked=False,
  155. )
  156. register(
  157. id="MiniGrid-ObstructedMaze-1Dlhb-v0",
  158. entry_point="gym_minigrid.envs.obstructedmaze:ObstructedMaze_1Dlhb",
  159. )
  160. register(
  161. id="MiniGrid-ObstructedMaze-2Dl-v0",
  162. entry_point="gym_minigrid.envs.obstructedmaze:ObstructedMaze_Full",
  163. agent_room=(2, 1),
  164. key_in_box=False,
  165. blocked=False,
  166. num_quarters=1,
  167. num_rooms_visited=4,
  168. )
  169. register(
  170. id="MiniGrid-ObstructedMaze-2Dlh-v0",
  171. entry_point="gym_minigrid.envs.obstructedmaze:ObstructedMaze_Full",
  172. agent_room=(2, 1),
  173. key_in_box=True,
  174. blocked=False,
  175. num_quarters=1,
  176. num_rooms_visited=4,
  177. )
  178. register(
  179. id="MiniGrid-ObstructedMaze-2Dlhb-v0",
  180. entry_point="gym_minigrid.envs.obstructedmaze:ObstructedMaze_Full",
  181. agent_room=(2, 1),
  182. key_in_box=True,
  183. blocked=True,
  184. num_quarters=1,
  185. num_rooms_visited=4,
  186. )
  187. register(
  188. id="MiniGrid-ObstructedMaze-1Q-v0",
  189. entry_point="gym_minigrid.envs.obstructedmaze:ObstructedMaze_Full",
  190. agent_room=(1, 1),
  191. key_in_box=True,
  192. blocked=True,
  193. num_quarters=1,
  194. num_rooms_visited=5,
  195. )
  196. register(
  197. id="MiniGrid-ObstructedMaze-2Q-v0",
  198. entry_point="gym_minigrid.envs.obstructedmaze:ObstructedMaze_Full",
  199. agent_room=(2, 1),
  200. key_in_box=True,
  201. blocked=True,
  202. num_quarters=2,
  203. num_rooms_visited=11,
  204. )
  205. register(
  206. id="MiniGrid-ObstructedMaze-Full-v0",
  207. entry_point="gym_minigrid.envs.obstructedmaze:ObstructedMaze_Full",
  208. )