leggedrobotics/terrain-grounded-imagination
Agent
Can I trust it?
Worth a closer look before you use it.
Warnings Check
Very new: it was created 4 days ago. Wait a little before installing it.
Stars from real accounts No data
We haven't checked its stars for fakes yet.
Active OK
Last change 2 days ago.
Licence OK
MIT: you can use it for free, also in a commercial product. Keep the licence notice.
Quick checks from public data. They help, but they are no guarantee.
What it is
Research code for training a Unitree G1 humanoid to walk over stairs, gaps and stepping stones using a terrain-conditioned world model with differentiable heightmap lookups. It includes an FPM foothold-prediction package for Isaac Lab 3 and a TGI package for world-model training…
Who it's for
Robotics and reinforcement learning researchers training legged locomotion policies in Isaac Lab
How to use it
Install
pip install -e fpm --no-depsFrom its README. Check it there before running it.
Ask your AI to set it up
Paste this into Claude, ChatGPT or your coding assistant.
Help me install and set up leggedrobotics/terrain-grounded-imagination in my project. Follow its README: https://github.com/leggedrobotics/terrain-grounded-imagination. Explain each step and check it is safe before running anything.
Star curve
Alternatives
Similar projects named in its profile. Each link searches the directory.
Get alerts for repos like this
We'll email you when something in Models & inference starts rising. Free, once a week, only what matches.
What arrives: one email on Mondays with the repos that match, never more. See last week's. Privacy
For maintainers
Add this badge to your README. It shows the repo's place on the radar, updates every hour and links back to this page.
[](https://repoalerts.com/repo/leggedrobotics/terrain-grounded-imagination?ref=badge)Similar repos
- apple-aiml-research/ml-halo
Research code for the HALO paper on perceptive humanoid motion learned from simulation. It covers synthetic data generation (Infinigen scenes, Kimodo motion, Isaac and MuJoCo rendering), training of a
- liuyisen-bin/transformer-sru-lys
Hierarchical reinforcement learning navigation for the Unitree G1 29DOF robot. A LiDAR CNN, spatial Transformer and SRU policy outputs velocity commands at 5 Hz for a frozen low-level controller at 50
- fuyichen0725/blinkmimic
Training code for whole-body motion tracking on the Unitree G1 humanoid. One policy is trained on all 40 LAFAN1 clips by backpropagating through rollouts of a learned world model, built on mjlab and M
- Huy-Quang-Dao/instantTrack
Training pipeline for whole-body motion tracking of the Unitree G1 humanoid, learning one policy for 40 LAFAN1 motion clips in a MuJoCo Warp simulator (mjlab). It builds on SuperMimic with a world mod
- zita-ch/supermimic_vibed
Unofficial reimplementation of the SuperMimic humanoid whole-body control method, built from the paper since no official code exists. Trains a single Unitree G1 motion-tracking policy on 40 LAFAN1 mot
- lzyang2000/morphmimic
A research codebase for training humanoid motion-tracking policies by backpropagating through a learned differentiable world model, with a single model shared across the Unitree G1, AgiBot X2 Ultra an