Fetch#


Action Space	`Discrete(7)`
Observation Space	`Dict('direction': Discrete(4), 'image': Box(0, 255, (7, 7, 3), uint8), 'mission': MissionSpace(<function FetchEnv._gen_mission at 0x7f813f518ee0>, [['get a', 'go get a', 'fetch a', 'go fetch a', 'you must fetch a'], ['blue', 'green', 'grey', 'purple', 'red', 'yellow'], ['key', 'ball']]))`
Reward Range	`(0, 1)`
Creation	`gymnasium.make("MiniGrid-Fetch-8x8-N3-v0")`

Description#

This environment has multiple objects of assorted types and colors. The agent receives a textual string as part of its observation telling it which object to pick up. Picking up the wrong object terminates the episode with zero reward.

Mission Space#

“{syntax} {color} {type}”

{syntax} is one of the following: “get a”, “go get a”, “fetch a”, “go fetch a”, “you must fetch a”.

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

{type} is the type of the object. Can be “key” or “ball”.

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	Unused
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 picks up the correct object.
The agent picks up the wrong object.
Timeout (see max_steps).

Registered Configurations#

N: number of objects to be generated.

MiniGrid-Fetch-5x5-N2-v0
MiniGrid-Fetch-6x6-N2-v0
MiniGrid-Fetch-8x8-N3-v0