Open Two Doors¶


Action Space	`Discrete(7)`
Observation Space	`Dict('direction': Discrete(4), 'image': Box(0, 255, (7, 7, 3), uint8), 'mission': MissionSpace(<function BabyAIMissionSpace._gen_mission at 0x7f756a28f0d0>, None))`
Creation	`gymnasium.make("BabyAI-OpenRedBlueDoorsDebug-v0")`

Description¶

Open door X, then open door Y The two doors are facing opposite directions, so that the agent Can’t see whether the door behind him is open. This task requires memory (recurrent policy) to be solved effectively.

Mission Space¶

“open the {color} door, then open the {color} door”

{color} is the color of the box. Can be “red”, “green”, “blue”, “purple”, “yellow” or “grey”.

Action Space¶

Num	Name	Action
0	left	Turn left
1	right	Turn right
2	forward	Move forward
3	pickup	Pick up an object
4	drop	Unused
5	toggle	Open/closes door
6	done	Unused

Observation Encoding¶

Each tile is encoded as a 3 dimensional tuple: (OBJECT_IDX, COLOR_IDX, STATE)
OBJECT_TO_IDX and COLOR_TO_IDX mapping can be found in minigrid/core/constants.py
STATE refers to the door state with 0=open, 1=closed and 2=locked

Rewards¶

A reward of ‘1 - 0.9 * (step_count / max_steps)’ is given for success, and ‘0’ for failure.

Termination¶

The episode ends if any one of the following conditions is met:

The agent opens the door.
Timeout (see max_steps).

Registered Configurations¶

BabyAI-OpenTwoDoors-v0
BabyAI-OpenRedBlueDoors-v0
BabyAI-OpenRedBlueDoorsDebug-v0