Surround#

../../../_images/atari_surround.gif

This environment is part of the Atari environments. Please read that page first for general information.

Import

from pettingzoo.atari import surround_v2

Actions

Discrete

Parallel API

Yes

Manual Control

No

Agents

agents= ['first_0', 'second_0']

Agents

2

Action Shape

(1,)

Action Values

[0,4]

Observation Shape

(210, 160, 3)

Observation Values

(0,255)

A competitive game of planning and strategy.

In surround, your goal is to avoid the walls. If you run into a wall, you are rewarded -1 points, and your opponent, +1 points.

But both players leave a trail of walls behind you, slowly filling the screen with obstacles. To avoid the obstacles as long as possible, you must plan your path to conserve space. Once that is mastered, a higher level aspect of the game comes into play, where both players literally try to surround the other with walls, so their opponent will run out of room and be forced to run into a wall.

Official surround manual

Environment parameters#

Environment parameters are common to all Atari environments and are described in the base Atari documentation .

Action Space (Minimal)#

In any given turn, an agent can choose from one of 6 actions. (Fire is dummy action, but for the continuous numbering)

Action

Behavior

0

No operation

1

Fire (dummy)

2

Move up

3

Move right

4

Move left

5

Move down

Version History#

  • v2: Minimal Action Space (1.18.0)

  • v1: Breaking changes to entire API (1.4.0)

  • v0: Initial versions release (1.0.0)

Usage#

AEC#

from pettingzoo.atari import surround_v2

env = surround_v2.env(render_mode="human")
env.reset(seed=42)

for agent in env.agent_iter():
    observation, reward, termination, truncation, info = env.last()

    if termination or truncation:
        action = None
    else:
        # this is where you would insert your policy
        action = env.action_space(agent).sample()

    env.step(action)
env.close()

Parallel#

from pettingzoo.atari import surround_v2

env = surround_v2.parallel_env(render_mode="human")
observations, infos = env.reset()

while env.agents:
    # this is where you would insert your policy
    actions = {agent: env.action_space(agent).sample() for agent in env.agents}

    observations, rewards, terminations, truncations, infos = env.step(actions)
env.close()

API#

class pettingzoo.atari.surround.surround.raw_env(**kwargs)[source]#