Skip to content

This project is meant to migrate core components and datasets of the C++ based UBC MOCCA software, which is called TerrainRLSim, into Gym.

Notifications You must be signed in to change notification settings

caoweiquan322/UbcMoccaMigration

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

UBC MOCCA Migration Project (Still in Progress)

This project is meant to migrate core components and datasets of the C++ based UBC MOCCA software, which is called (TerrainRLSim)[https://github.com/UBCMOCCA/TerrainRLSim], into Gym. Thus this useful tool would be much easier to use, and potentially attract even more researchers join it.

Roadmap

Done

TODO by now

  • Handle joint limits/range.
  • Add terrain data.
  • Load and play motion data.

In progress

  • Converting MOCCA terrain (JSON format) into mujoco format.
  • Parsing MOCCA motion files with python.
  • Play motion data with mujoco based environment.
  • Being able to access motion state, e.g. position, quaternion, velocity etc., of every body part and end effector easily.
  • Creating learning tasks (or known as env) with Gym manner.
  • Creating RL learning nets, or reusing open source implementations such as (OpenAI)[https://openai.github.io] and DeepMind.

Python version&requirements

Python version 3.6+ is recommended.

Besides the requirements declared in requirements.txt, some external packages are required.

About

This project is meant to migrate core components and datasets of the C++ based UBC MOCCA software, which is called TerrainRLSim, into Gym.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages