RoomObjects#


Action Space	`Discrete(8)`
Observation Space	`Box(0, 255, (60, 80, 3), uint8)`
Reward Range	`(-inf, inf)`
Creation	`gymnasium.make("MiniWorld-RoomObjects-v0")`

Description#

Single room with multiple objects. Inspired by the single room environment of the Generative Query Networks paper: https://deepmind.com/blog/neural-scene-representation-and-rendering/

Action Space#

Num	Action
0	turn left
1	turn right
2	move forward
3	move back
4	pick up
5	drop
6	toggle / activate an object
7	complete task

Observation Space#

The observation space is an ndarray with shape (obs_height, obs_width, 3) representing a RGB image of what the agents sees.

Rewards:#

None

Arguments#

RoomObjects(size=16)

size: size of world