Skip to content

Reproducing MuJoCo benchmarks in a modern, commercial game /physics engine (Unity + PhysX).

License

Notifications You must be signed in to change notification settings

Sohojoe/MujocoUnity

Repository files navigation

MujocoUnity

Reproducing MuJoCo benchmarks in a modern, commercial game /physics engine (Unity + PhysX).

Presented March 19th, 2018 at the AI Summit - Game Developer Conference 2018 - http://schedule.gdconf.com/session/beyond-bots-making-machine-learning-accessible-and-useful/856147


IMPORTANT: Active development has moved to Marathon-envs

MujocoUnity


Legacy readme...

v0.2

supports Unity ml-agents (instructions)

Trained with ml-agents PPO:

unity_oai_hopper.xml

Mujoco Hopper 1.5m 1.5m steps

unity_dm_walker.xml

Mujoco Hopper 1.5m 3m steps

Trained with Baselines DDPG:

unity_oai_hopper.xml

Mujoco Hopper 300k 300k steps Mujoco Hopper 300k 2m steps

unity_dm_walker.xml

Mujoco Hopper 300k 1m steps Mujoco Hopper 300k 3m steps

Note: to reproduce you'll need to figure out how to patch OpenAI baselines with Unity.

Known Issues:

  • oai_humanoid - is broken. Configurable joint needs updating to support multi-directional joints
  • oai_half_cheetah - need to implement geom axis-angle
  • dm_xxxx - need to implement class=
  • phyx is not tuned properly
  • Capsules are slightly too long - NOTE: this can cause collision issues whereby the leg may slightly poke the a foot and trigger collisions

From earlier version:

Ant - Random Mujoco Ant random policy Ant - trained with ACKTR Mujoco Ant trainged with acktr