ti5robot-15.53

Name: ti5robot-15.53 Version: 1.0.0 Summary: mujoco environments for humanoid robot Author: Huiyang Kong Author-email: huiyang.kong@ti5robot.com License: BSD-3-Clause

Code for the papers:

• Learning Bipedal Walking On Planned Footsteps For Humanoid Robots (Humanoids2022) Rohan P. Singh, Mehdi Benallegue, Mitsuharu Morisawa, Rafael Cisneros, Fumio Kanehiro

• Learning Bipedal Walking for Humanoids with Current Feedback (arxiv) Rohan P. Singh, Zhaoming Xie, Pierre Gergondet, Fumio Kanehiro (WIP on branch topic/omnidirectional-walk)

Code structure:

A rough outline for the repository that might be useful for adding your own robot:

ti5robot-15.53/
├── envs/                <-- Actions and observation space, PD gains, simulation step, control decimation, init, ...
├── tasks/               <-- Reward function, termination conditions, and more...
├── rl/                  <-- Code for PPO, actor/critic networks, observation normalization process...
├── models/              <-- MuJoCo model files: XMLs/meshes/textures
├── trained/             <-- Contains pretrained model for JVRC
└── scripts/             <-- Utility scripts, etc.

Requirements:

• Python version: 3.10.6
• Pytorch
• pip install:
  • mujoco==3.2.6
  • mujoco-python-viewer==0.1.4
  • ray==2.40.0
  • transforms3d==0.4.2
  • matplotlib==3.9.3
  • scipy==1.14.1
  • torch==2.5.1

Usage:

Environment names supported:

Task Description	Environment name
Basic Walking Task	'jvrc_walk'
Stepping Task (using footsteps)	'jvrc_step'

To train:

$ python run_experiment.py train --logdir <path_to_exp_dir> --num_procs <num_of_cpu_procs> --env <name_of_environment>

To play:

$ PYTHONPATH=.:$PYTHONPATH python scripts/debug_stepper.py --path <path_to_exp_dir>

What you should see:

Ascending stairs:

Descending stairs:

Walking on curves: