Humanoid-Gym: Reinforcement Learning for Humanoid Robot with Zero-Shot Sim2Real Transfer

Humanoid-Gym is an easy-to-use reinforcement learning (RL) framework based on Nvidia Isaac Gym, designed to train locomotion skills for humanoid robots, emphasizing zero-shot transfer from simulation to the real-world environment. Humanoid-Gym also integrates a sim-to-sim framework from Isaac Gym to Mujoco that allows users to verify the trained policies in different physical simulations to ensure the robustness and generalization of the policies.

This codebase is verified by RobotEra's XBot-S (1.2 meter tall humanoid robot) and XBot-L (1.65 meter tall humanoid robot) in real-world environment with zero-shot sim-to-real transfer.

Robot Hardware Platform for Evaluation

XBot-S

Tall: 1.2 meters

Weight: 38 kg

Actuated Motors: 26

Total DOF: 32

XBot-L

Tall: 1.65 meters

Weight: 57 kg

Actuated Motors: 54

Total DOF: 60

Performance Comparison in Different Environments

isaac_gym.mp4

Nvidia Isaac Gym

mujoco-plane.mp4

Sim-to-sim (Mujoco)

real-world.mp4

Sim-to-real

Challenging Terrains With Zero-shot Sim-to-sim Transfer

mujoco-terrain-front-view.mov

Front View

mujoco-terrain-side-view.mov

Side View

Challenging Locomotion Task in Real-World Environments (Denoising World Model Learning)

snowy-terrain.mp4

Snowy Terrain

uneven-terrain.mp4

Dynamic Uneven Terrain

dynamic-payload.mp4

Dynamic Payload

(10 kg)

external-force.mp4

External Force

stairs.mov

Go Up/Down Stairs

push-60kg.mp4

Push 60 kg Cart

uneven-terrain-ankle-view.mp4

Uneven Terrain 

(Ankle View)

slopes.mp4

Go Up/Down Slopes

Dexterous Hand Manipulation (Coming Soon)

stack-blocks.mp4

Stack Blocks

dexterous-manipulation-prepare-food.mp4

Prepare Food

Citation


@article{gu2024humanoid,

  title={Humanoid-Gym: Reinforcement Learning for Humanoid Robot with Zero-Shot Sim2Real Transfer},

  author={Gu, Xinyang and Wang, Yen-Jen and Chen, Jianyu},

  journal={arXiv preprint arXiv:2404.05695},

  year={2024}

}

Acknowledgment

The implementation of Humanoid-Gym relies on resources from legged_gym and rsl_rl projects, created by the Robotic Systems Lab. We specifically utilize the LeggedRobot implementation from their research to enhance our codebase.

Any Questions?

If you have further questions, please feel free to contact support@robotera.com.