Unify Isaac Lab assets with Isaac Sim #4301

matthewtrepte · 2025-12-29T22:38:55Z

Description

Rebasing and benchmarking new assets from Steven Feng's unify assets change - #3608

Move to use new USD variants for G1 and H1 assets.

Type of change

New feature (non-breaking change which adds functionality)

Screenshots

Please attach before and after screenshots of the change if applicable.

Checklist

I have read and understood the contribution guidelines
I have run the pre-commit checks with ./isaaclab.sh --format
I have made corresponding changes to the documentation
My changes generate no new warnings
I have added tests that prove my fix is effective or that my feature works
I have updated the changelog and the corresponding version in the extension's config/extension.toml file
I have added my name to the CONTRIBUTORS.md or my name already exists there

- Update test_utils_queries.py to use ISAAC_NUCLEUS_DIR with correct paths - Fix G1 Inspire Hand path to use ISAAC_NUCLEUS_DIR/Robots/Unitree/G1_29dof/ - Remove unused stage_utils imports from test files - Ensure consistent path naming (ANYmal-C -> anymal_c, etc.)

greptile-apps · 2025-12-29T22:42:16Z

Greptile Summary

Migrated Isaac Lab asset paths from ISAACLAB_NUCLEUS_DIR to ISAAC_NUCLEUS_DIR to unify with Isaac Sim's asset structure. This changes all robot USD paths to reference the shared Isaac Sim asset repository on the Nucleus server.

Key Changes:

All robot configurations now use ISAAC_NUCLEUS_DIR (Isaac/Robots/) instead of ISAACLAB_NUCLEUS_DIR (Isaac/IsaacLab/Robots/)
Updated directory structures for ANYmal variants (ANYmal-B → anymal_b, ANYmal-C → anymal_c, ANYmal-D → anymal_d)
Franka Panda path updated to Robots/FrankaRobotics/FrankaEmika/
Unitree G1 paths updated to specify DOF variants (G1 → G1_23dof, G1_29dof)
Humanoid28 moved to Robots/IsaacSim/Humanoid28/
All test files, demos, and task environments updated consistently
Removed unused ISAACLAB_NUCLEUS_DIR imports where no longer needed

Asset Path Pattern:

Before: {ISAACLAB_NUCLEUS_DIR}/Robots/[Manufacturer]/[Model]/[file].usd
After: {ISAAC_NUCLEUS_DIR}/Robots/[Manufacturer]/[model]/[file].usd

The migration is systematic and comprehensive, touching all robot assets, examples, and tests.

Confidence Score: 4/5

This PR is safe to merge after verifying that the new asset paths exist on the Nucleus server
The changes are purely path migrations with no logic modifications. All updates are consistent and systematic. Score is 4 (not 5) because this depends on the corresponding assets being available at the new paths on the Nucleus server, which cannot be verified from the code alone
Verify that all new asset paths exist on the Nucleus server before merging, particularly the reorganized paths for ANYmal variants, G1 DOF-specific paths, and the FrankaRobotics directory structure

Important Files Changed

Filename	Overview
source/isaaclab/isaaclab/sim/spawners/init.py	Updated documentation examples to use new ISAAC_NUCLEUS_DIR paths instead of ISAACLAB_NUCLEUS_DIR
source/isaaclab_assets/isaaclab_assets/robots/franka.py	Migrated Franka Panda asset path from ISAACLAB_NUCLEUS_DIR to ISAAC_NUCLEUS_DIR/Robots/FrankaRobotics/FrankaEmika
source/isaaclab_assets/isaaclab_assets/robots/anymal.py	Migrated all ANYmal variants (B, C, D) to ISAAC_NUCLEUS_DIR with updated directory structure (lowercase names)
source/isaaclab_assets/isaaclab_assets/robots/unitree.py	Migrated all Unitree robots (A1, Go1, Go2, H1, G1, G1_29DOF) to ISAAC_NUCLEUS_DIR with updated paths for G1 variants
source/isaaclab/test/sim/test_spawn_from_files.py	Updated test cases to use new ISAAC_NUCLEUS_DIR path for Franka Panda robot
source/isaaclab/test/deps/isaacsim/check_legged_robot_clone.py	Unified isaaclab and oige asset paths to both use ISAAC_NUCLEUS_DIR for ANYmal-C

Sequence Diagram

sequenceDiagram
    participant User as User Code
    participant Config as Robot Config
    participant Assets as isaaclab.utils.assets
    participant Spawner as sim_utils.UsdFileCfg
    participant NucleusServer as Nucleus Server
    participant Stage as USD Stage

    User->>Config: Import robot configuration
    Config->>Assets: Use ISAAC_NUCLEUS_DIR constant
    Note over Assets: ISAAC_NUCLEUS_DIR = carb.settings<br/>/persistent/isaac/asset_root/cloud/Isaac
    Config->>Config: Build asset path<br/>(e.g., {ISAAC_NUCLEUS_DIR}/Robots/FrankaRobotics/FrankaEmika/panda.usd)
    User->>Spawner: Create UsdFileCfg with usd_path
    Spawner->>Assets: check_file_path(usd_path)
    Assets->>NucleusServer: omni.client.stat(usd_path)
    NucleusServer-->>Assets: Return file status
    Assets-->>Spawner: File exists on Nucleus
    Spawner->>NucleusServer: Load USD from remote path
    NucleusServer-->>Spawner: USD file data
    Spawner->>Stage: Spawn prim at prim_path
    Stage-->>User: Robot instantiated

greptile-apps · 2025-12-29T22:42:17Z

Greptile's behavior is changing!

From now on, if a review finishes with no comments, we will not post an additional "statistics" comment to confirm that our review found nothing to comment on. However, you can confirm that we reviewed your changes in the status check section.

_{This feature can be toggled off in your Code Review Settings by deselecting "Create a status check for each PR".}

pascal-roth · 2026-01-09T15:35:45Z

what is the status with the combination of the assets?

matthewtrepte · 2026-01-13T00:09:05Z

what is the status with the combination of the assets?

env tests are passing, spotchecked some standalone scripts which all work, and ran a full batch of training benchmarks for all envs for all workflows

one benchmark regression spotted so far is w the g1 29 DOF asset, so following up on that at the moment

source/Untitled

scripts/demos/multi_asset.py

source/isaaclab_assets/isaaclab_assets/robots/cassie.py

source/isaaclab_assets/isaaclab_assets/robots/unitree.py

kellyguo11 · 2026-01-13T01:05:04Z

source/isaaclab_tasks/test/test_environments.py



-@pytest.mark.parametrize("num_envs, device", [(32, "cuda"), (1, "cuda")])
+# @pytest.mark.parametrize("num_envs, device", [(32, "cuda"), (1, "cuda")])


why is this limited to 1 num_env?

@matthewtrepte

stevfeng · 2026-01-13T17:26:14Z

source/isaaclab_assets/isaaclab_assets/robots/unitree.py

 G1_INSPIRE_FTP_CFG = G1_29DOF_CFG.copy()
-G1_INSPIRE_FTP_CFG.spawn.usd_path = f"{ISAACLAB_NUCLEUS_DIR}/Robots/Unitree/G1/g1_29dof_inspire_hand.usd"
+G1_INSPIRE_FTP_CFG.spawn.usd_path = f"{ISAAC_NUCLEUS_DIR}/Robots/Unitree/G1/g1.usd"
+G1_INSPIRE_FTP_CFG.spawn.variants = {"Physics": "PhysX", "right_hand": "Inspire", "left_hand": "Inspire"}


@matthewtrepte This looks correct, we also have a SimplifiedPhysX for just feet, calf, torso, head, hand colliders

okay, I'll reach out to Peter Du to see if there's a preference

stevfeng and others added 9 commits December 29, 2025 22:10

update anymal, unitree, cartpole, cassie paths

b24edab

use internal path, update unitree

383adfb

update robot paths

dc1c1f3

format code

84f6b14

update cassie to use temp asset for testing before we merge it in

cadb828

revert cassie path

a6f303d

update asset path to staging server

0b22645

update path for the usd test

adb8286

matthewtrepte requested review from Mayankm96, jtigue-bdai, kellyguo11, ooctipus and pascal-roth as code owners December 29, 2025 22:38

github-actions bot added the asset New asset feature or request label Dec 29, 2025

matthewtrepte added 4 commits January 5, 2026 19:04

update cartpole path

b0cd153

change gl_29dof

06f641f

Merge branch 'main' into mtrepte/unify_assets

c94d81d

include recent update to assets

9c8d980

Mayankm96 added this to Isaac Lab Jan 12, 2026

Mayankm96 moved this to In progress in Isaac Lab Jan 12, 2026

matthewtrepte added 2 commits January 12, 2026 18:04

update usd variants for G1 (for feedback)

f47918b

updating g1 asset paths

c79a829

kellyguo11 reviewed Jan 13, 2026

View reviewed changes

stevfeng reviewed Jan 13, 2026

View reviewed changes

clean up, revert g1/h1 assets

672a1b3

Mayankm96 mentioned this pull request Jan 15, 2026

Unify Isaac Lab assets with Isaac Sim #3608

Closed

7 tasks

matthewtrepte added 2 commits January 15, 2026 21:46

merge

8fc7e7a

formatting

543de41

matthewtrepte changed the title ~~[WIP] Unify Isaac Lab assets with Isaac Sim~~ Unify Isaac Lab assets with Isaac Sim Jan 15, 2026

matthewtrepte requested review from Toni-SM, hhansen-bdai, masoudmoghani, ossamaAhmed, peterd-NV and xyao-nv as code owners January 16, 2026 18:56

github-actions bot added documentation Improvements or additions to documentation isaac-mimic Related to Isaac Mimic team infrastructure labels Jan 16, 2026

matthewtrepte force-pushed the mtrepte/unify_assets branch from 7fdb303 to 543de41 Compare January 16, 2026 18:58

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Unify Isaac Lab assets with Isaac Sim #4301

Unify Isaac Lab assets with Isaac Sim #4301

matthewtrepte commented Dec 29, 2025 •

edited

Loading

Uh oh!

greptile-apps bot commented Dec 29, 2025

Uh oh!

greptile-apps bot commented Dec 29, 2025

Uh oh!

pascal-roth commented Jan 9, 2026

Uh oh!

matthewtrepte commented Jan 13, 2026 •

edited

Loading

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

kellyguo11 Jan 13, 2026

Uh oh!

stevfeng Jan 13, 2026

Uh oh!

stevfeng Jan 13, 2026

Uh oh!

matthewtrepte Jan 15, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants



		@pytest.mark.parametrize("num_envs, device", [(32, "cuda"), (1, "cuda")])
		# @pytest.mark.parametrize("num_envs, device", [(32, "cuda"), (1, "cuda")])

Unify Isaac Lab assets with Isaac Sim #4301

Are you sure you want to change the base?

Unify Isaac Lab assets with Isaac Sim #4301

Conversation

matthewtrepte commented Dec 29, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Type of change

Screenshots

Checklist

Uh oh!

greptile-apps bot commented Dec 29, 2025

Greptile Summary

Confidence Score: 4/5

Important Files Changed

Sequence Diagram

Uh oh!

greptile-apps bot commented Dec 29, 2025

Greptile's behavior is changing!

Uh oh!

pascal-roth commented Jan 9, 2026

Uh oh!

matthewtrepte commented Jan 13, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

kellyguo11 Jan 13, 2026

Choose a reason for hiding this comment

Uh oh!

stevfeng Jan 13, 2026

Choose a reason for hiding this comment

Uh oh!

stevfeng Jan 13, 2026

Choose a reason for hiding this comment

Uh oh!

matthewtrepte Jan 15, 2026

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

matthewtrepte commented Dec 29, 2025 •

edited

Loading

matthewtrepte commented Jan 13, 2026 •

edited

Loading