Go To Object#

Go To Object

Action Space

Discrete(7)

Observation Space

Dict('direction': Discrete(4), 'image': Box(0, 255, (7, 7, 3), uint8), 'mission': MissionSpace(<function GoToObjectEnv._gen_mission at 0x7f813f507790>, [['blue', 'green', 'grey', 'purple', 'red', 'yellow'], ['key', 'ball', 'box']]))

Reward Range

(0, 1)

Creation

gymnasium.make("MiniGrid-GoToObject-8x8-N2-v0")

Description#

This environment is a room with colored objects. The agent receives a textual (mission) string as input, telling it which colored object to go to, (eg: “go to the red key”). It receives a positive reward for performing the done action next to the correct object, as indicated in the mission string.

Mission Space#

“go to the {color} {obj_type}”

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

Action Space#

Num

Name

Action

0

left

Turn left

1

right

Turn right

2

forward

Move forward

3

pickup

Unused

4

drop

Unused

5

toggle

Unused

6

done

Done completing task

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:

  1. The agent stands next the correct door performing the done action.

  2. Timeout (see max_steps).

Registered Configurations#

  • MiniGrid-GoToObject-6x6-N2-v0

  • MiniGrid-GoToObject-8x8-N2-v0