https://github.com/geyang/legged-control-suite

A collection of legged robot environments built off DeepMind control
https://github.com/geyang/legged-control-suite

Last synced: about 2 months ago
JSON representation

A collection of legged robot environments built off DeepMind control

• Host: GitHub
• URL: https://github.com/geyang/legged-control-suite
• Owner: geyang
• License: mit
• Created: 2023-01-13T04:50:42.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2024-03-08T23:19:29.000Z (10 months ago)
• Last Synced: 2024-04-14T15:11:51.291Z (9 months ago)
• Language: Python
• Size: 1.89 MB
• Stars: 3
• Watchers: 2
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# Legged Control Suite

A collection of legged robot environments built upon the DeepMind control

suite and MuJoCo physics engine.

## Environments

- Hopper (3D)

- Bipedal (3D)

- Agile Bipedal Robot (ours)

## To Do

As excersize, fork this repo and hack around the walker environment and 

  make a simple bipedal walker. The `lcs` module contains two walker 

  scripts copied from DeepMind Control Suite. You need to:

- [ ] First be able to load `lcs:BipedalWalker-v0`. This is currently not 

      runnable.

- [ ] Change the name to `BipedalWalker` and add the following tasks: `["walk",

      "run", "hop"]`