MO-Halfcheetah

../../_images/mo-halfcheetah.gif

Action Space

Box(-1.0, 1.0, (6,), float32)

Observation Shape

(17,)

Observation High

inf

Observation Low

-inf

Reward Shape

(2,)

Reward High

[inf inf]

Reward Low

[-inf -inf]

Import

mo_gymnasium.make("mo-halfcheetah-v5")

Description

Multi-objective version of the HalfCheetahEnv environment.

See Gymnasium’s env for more information.

The original Gymnasium’s ‘HalfCheetah-v5’ is recovered by the following linear scalarization:

env = mo_gym.make(‘mo-halfcheetah-v5’) LinearReward(env, weight=np.array([1.0, 0.1]))

Reward Space

The reward is 2-dimensional:

  • 0: Reward for running forward

  • 1: Control cost of the action

Version History

  • v5: The scales of the control cost has changed from v4. See https://gymnasium.farama.org/environments/mujoco/half_cheetah/#version-history for other changes.