google-deepmind / google-deepmind/mujoco_playground

Pretrained g1 policy doesn't seem to respect command well

Open
#77 6 comments 0 reactions 1 assignee Claimed by @kevinzakka View on GitHub
enhancement
Dominant language
Python
Stars
2.2k
Forks
359
Avg merge
7d 3h
Merged PRs (30d)
1

Description

I tried the play_g1_joystick.py example with the default g1 onnx policy, without connecting it to a gamepad so the command is always [0, 0, 0]. Seems like the robot still moves forward and doesn't track orientation well. Is this expected?

https://github.com/user-attachments/assets/7b535b02-8d63-4d88-9f44-1eb1bd12f4ad

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.