<h1 align="center">Human to Humanoid</h1>

Official Implementation for H2O and OmniH2O:

Learning Human-to-Humanoid Real-Time Whole-Body Teleoperation, IROS 2024.
OmniH2O: Universal and Dexterous Human-to-Humanoid Whole-Body Teleoperation and Learning, CoRL 2024.

This codebase is under CC BY-NC 4.0 license, with inherited license in Legged Gym and RSL RL from ETH Zurich, Nikita Rudin and NVIDIA CORPORATION & AFFILIATES. You may not use the material for commercial purposes, e.g., to make demos to advertise your commercial products.

Please read through the whole README.md before cloning the repo.

Installation

Note: Before running our code, it's highly recommended to first play with RSL's Legged Gym version to get a basic understanding of the Isaac-LeggedGym-RslRL framework.

Create environment and install torch

conda create -n omnih2o python=3.8 
conda activate omnih2o
pip3 install torch torchvision torchaudio

Install Isaac Gym preview 4 release https://developer.nvidia.com/isaac-gym

unzip files to a folder, then install with pip:

cd isaacgym/python && pip install -e .

check it is correctly installed by playing:
```
cd examples && python 1080_balls_of_solitude.py
```
Clone this codebase and install our rsl_rl in the training folder
```
pip install -e rsl_rl
```
Install our legged_gym
```
pip install -e legged_gym
```
Ensure you have installed the following packages:
- pip install numpy==1.20 (must < 1.24)
- pip install tensorboard
- pip install setuptools==59.5.0
Install our phc
```
pip install -e phc
```
Install additional packages requirements.txt
```
pip install -r requirements.txt
```

Training and Playing

Try training and playing privileged teacher policy.

can use "--headless" to disable gui, press "v" to pause/resume gui play.

# OmniH2O Training and Playing Teacher Policy 
python legged_gym/scripts/train_hydra.py --config-name=config_teleop task=h1:teleop run_name=OmniH2O_TEACHER env.num_observations=913 env.num_privileged_obs=990 motion.teleop_obs_version=v-teleop-extend-max-full motion=motion_full motion.extend_head=True num_envs=4096 asset.zero_out_far=False asset.termination_scales.max_ref_motion_distance=1.5 sim_device=cuda:0 motion.motion_file=resources/motions/h1/stable_punch.pkl rewards=rewards_teleop_omnih2o_teacher rewards.penalty_curriculum=True rewards.penalty_scale=0.5

# OmniH2O Play Teacher Policy
python  legged_gym/scripts/play_hydra.py --config-name=config_teleop task=h1:teleop env.num_observations=913 env.num_privileged_obs=990 motion.future_tracks=True motion.teleop_obs_version=v-teleop-extend-max-full motion=motion_full  motion.extend_head=True asset.zero_out_far=False asset.termination_scales.max_ref_motion_distance=10.0  sim_device=cuda:0 load_run=OmniH2O_TEACHER checkpoint=XXXX num_envs=1 headless=False

Try training and playing sim2real deploy policy.

# OmniH2O Distill Student Policy
python legged_gym/scripts/train_hydra.py --config-name=config_teleop task=h1:teleop run_name=OmniH2O_STUDENT env.num_observations=1665 env.num_privileged_obs=1742 motion.teleop_obs_version=v-teleop-extend-vr-max-nolinvel motion.teleop_selected_keypoints_names=[] motion.extend_head=True num_envs=4096 asset.zero_out_far=False asset.termination_scales.max_ref_motion_distance=1.5 sim_device=cuda:0 motion.motion_file=resources/motions/h1/stable_punch.pkl rewards=rewards_teleop_omnih2o_teacher rewards.penalty_curriculum=True rewards.penalty_scale=0.5 train.distill=True train.policy.init_noise_std=0.001 env.add_short_history=True env.short_history_length=25 noise.add_noise=False noise.noise_level=0 train.dagger.load_run_dagger=TEACHER_RUN_NAME train.dagger.checkpoint_dagger=XXX train.dagger.dagger_only=True

# OmniH2O Play Student Policy
python legged_gym/scripts/play_hydra.py --config-name=config_teleop task=h1:teleop env.num_observations=1665 env.num_privileged_obs=1742 motion.teleop_obs_version=v-teleop-extend-vr-max-nolinvel motion.teleop_selected_keypoints_names=[] motion.extend_head=True num_envs=1 asset.zero_out_far=False asset.termination_scales.max_ref_motion_distance=10.0 sim_device=cuda:0 load_run=OmniH2O_STUDENT checkpoint=XXXX env.add_short_history=True env.short_history_length=25 headless=False

Different Configurations on Hisotry Steps

0-step MLP

# OmniH2O Distill 0-step MLP Student Policy 
python legged_gym/scripts/train_hydra.py --config-name=config_teleop task=h1:teleop run_name=OmniH2O_STUDENT_0stepMLP env.num_observations=90 env.num_privileged_obs=167 motion.teleop_obs_version=v-teleop-extend-vr-max-nolinvel motion.teleop_selected_keypoints_names=[] motion.extend_head=True num_envs=4096 asset.zero_out_far=False asset.termination_scales.max_ref_motion_distance=1.5 sim_device=cuda:0 motion.motion_file=resources/motions/h1/stable_punch.pkl rewards=rewards_teleop_omnih2o_teacher rewards.penalty_curriculum=True rewards.penalty_scale=0.5 train.distill=True train.policy.init_noise_std=0.001 env.add_short_history=False noise.add_noise=False noise.noise_level=0 train.dagger.load_run_dagger=TEACHER_RUN_NAME train.dagger.checkpoint_dagger=XXX train.dagger.dagger_only=True

# OmniH2O Play 0-step MLP Student Policy 
python legged_gym/scripts/play_hydra.py --config-name=config_teleop task=h1:teleop env.num_observations=90 env.num_privileged_obs=167 motion.teleop_obs_version=v-teleop-extend-vr-max-nolinvel motion.teleop_selected_keypoints_names=[] motion.extend_head=True num_envs=1 asset.zero_out_far=False asset.termination_scales.max_ref_motion_distance=10.0 sim_device=cuda:0 load_run=OmniH2O_STUDENT checkpoint=XXXX env.add_short_history=False headless=False

5-step MLP

# OmniH2O Distill 5-step MLP Student Policy 
python legged_gym/scripts/train_hydra.py --config-name=config_teleop task=h1:teleop run_name=OmniH2O_STUDENT_50stepMLP env.num_observations=405 env.num_privileged_obs=482 motion.teleop_obs_version=v-teleop-extend-vr-max-nolinvel motion.teleop_selected_keypoints_names=[] motion.extend_head=True num_envs=4096 asset.zero_out_far=False asset.termination_scales.max_ref_motion_distance=1.5 sim_device=cuda:0 motion.motion_file=resources/motions/h1/stable_punch.pkl rewards=rewards_teleop_omnih2o_teacher rewards.penalty_curriculum=True rewards.penalty_scale=0.5 train.distill=True train.policy.init_noise_std=0.001 env.add_short_history=True env.short_history_length=5 noise.add_noise=False noise.noise_level=0 train.dagger.load_run_dagger=TEACHER_RUN_NAME train.dagger.checkpoint_dagger=XXX train.dagger.dagger_only=True

# OmniH2O Play 5-step MLP Student Policy 
python legged_gym/scripts/play_hydra.py --config-name=config_teleop task=h1:teleop env.env.env.num_observations=405 env.num_privileged_obs=482 motion.teleop_obs_version=v-teleop-extend-vr-max-nolinvel motion.teleop_selected_keypoints_names=[] motion.extend_head=True num_envs=1 asset.zero_out_far=False asset.termination_scales.max_ref_motion_distance=10.0 sim_device=cuda:0 load_run=OmniH2O_STUDENT checkpoint=XXXX env.add_short_history=True env.short_history_length=5 headless=False

50-step MLP

# OmniH2O Distill 50-step MLP Student Policy 
python legged_gym/scripts/train_hydra.py --config-name=config_teleop task=h1:teleop run_name=OmniH2O_STUDENT_50stepMLP env.num_observations=3240 env.num_privileged_obs=3317 motion.teleop_obs_version=v-teleop-extend-vr-max-nolinvel motion.teleop_selected_keypoints_names=[] motion.extend_head=True num_envs=4096 asset.zero_out_far=False asset.termination_scales.max_ref_motion_distance=1.5 sim_device=cuda:0 motion.motion_file=resources/motions/h1/stable_punch.pkl rewards=rewards_teleop_omnih2o_teacher rewards.penalty_curriculum=True rewards.penalty_scale=0.5 train.distill=True train.policy.init_noise_std=0.001 env.add_short_history=True env.short_history_length=50 noise.add_noise=False noise.noise_level=0 train.dagger.load_run_dagger=TEACHER_RUN_NAME train.dagger.checkpoint_dagger=XXX train.dagger.dagger_only=True

# OmniH2O Play 50-step MLP Student Policy 
python legged_gym/scripts/play_hydra.py --config-name=config_teleop task=h1:teleop env.env.num_observations=3240 env.num_privileged_obs=3317 motion.teleop_obs_version=v-teleop-extend-vr-max-nolinvel motion.teleop_selected_keypoints_names=[] motion.extend_head=True num_envs=1 asset.zero_out_far=False asset.termination_scales.max_ref_motion_distance=10.0 sim_device=cuda:0 load_run=OmniH2O_STUDENT checkpoint=XXXX env.add_short_history=True env.short_history_length=50 headless=False

Different Configurations on Hisotry Architectures

LSTM

# OmniH2O Distill LSTM Student Policy 
python legged_gym/scripts/train_hydra.py --config-name=config_teleop task=h1:teleop run_name=LSTM_STUDENT env.num_observations=90 env.num_privileged_obs=167 motion.teleop_obs_version=v-teleop-extend-vr-max-nolinvel motion.teleop_selected_keypoints_names=[] motion.extend_head=True num_envs=4096 asset.zero_out_far=False asset.termination_scales.max_ref_motion_distance=1.5 sim_device=cuda:0 motion.motion_file=resources/motions/h1/stable_punch.pkl rewards=rewards_teleop_omnih2o_teacher rewards.penalty_curriculum=False rewards.penalty_scale=0.5 train.distill=True train

Human2humanoid

Install / Use

README

Installation

Training and Playing