MO-Humanoid¶
Action Space |
Box(-0.4, 0.4, (17,), float32) |
Observation Shape |
(348,) |
Observation High |
inf |
Observation Low |
-inf |
Reward Shape |
(2,) |
Reward High |
[inf inf] |
Reward Low |
[-inf -inf] |
Import |
|
Description¶
Multi-objective version of the HumanoidEnv environment.
See Gymnasium’s env for more information.
The original Gymnasium’s ‘Humanoid-v5’ is recovered by the following linear scalarization:
env = mo_gym.make(‘mo-humanoid-v5’) LinearReward(env, weight=np.array([1.25, 0.1]))
Reward Space¶
The reward is 2-dimensional:
0: Reward for running forward (x-velocity)
1: Control cost of the action
Version History:¶
v5: Now includes contact forces. See: https://gymnasium.farama.org/environments/mujoco/humanoid/#version-history The scales of the control cost has changed from v4.