Simple

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

Import	from pettingzoo.mpe import simple_v3
Actions	Discrete/Continuous
Parallel API	Yes
Manual Control	No
Agents	agents= [agent_0]
Agents	1
Action Shape	(5)
Action Values	Discrete(5)/Box(0.0, 1.0, (5,))
Observation Shape	(4)
Observation Values	(-inf,inf)
State Shape	(4,)
State Values	(-inf,inf)

In this environment a single agent sees a landmark position and is rewarded based on how close it gets to the landmark (Euclidean distance). This is not a multiagent environment, and is primarily intended for debugging purposes.

Observation space: [self_vel, landmark_rel_position]

Arguments

simple_v3.env(max_cycles=25, continuous_actions=False)

max_cycles: number of frames (a step for each agent) until game terminates

continuous_actions: Whether agent action spaces are discrete(default) or continuous

API

class pettingzoo.mpe.simple.simple.raw_env(max_cycles=25, continuous_actions=False, render_mode=None)

Uses the args and kwargs from the object’s constructor for pickling.

action_spaces: dict[AgentID, gymnasium.spaces.Space]

agent_selection: AgentID

agents: list[AgentID]

infos: dict[AgentID, dict[str, Any]]

observation_spaces: dict[AgentID, gymnasium.spaces.Space]

possible_agents: list[AgentID]

rewards: dict[AgentID, float]

terminations: dict[AgentID, bool]

truncations: dict[AgentID, bool]