Repository navigation
Tutorials/Sim2Real Blogs: Add MuJoCo and MJWarp robot lessons. - #279
Merged
Merged
Conversation
Add Blog 2 companion notebooks, exercises, solutions, reproducible dependencies, and task/throughput validation. Preserve source and asset attribution. Signed-off-by: Johnny <johnnynuca14@gmail.com>
Contributor
|
👋 Thank you for your contribution! This pull request is from a forked repository so GitHub Actions will not be able to run CI. A maintainer will review your changes shortly and manually trigger the CI. @maintainers Please review this PR when you have a chance and follow the instructions in the CONTRIBUTING.md file to trigger the CI. |
Explain that the scripted IK uses command-derived scratch state, while copied GPU state supports rendering and task inspection. Signed-off-by: Johnny <johnnynuca14@gmail.com>
Restore the original Warp tutorial and move the SO-101 lessons, tests, and Brev environment into tutorials/sim2real-blogs. Update notebook numbering, setup paths, navigation, and the tutorial index. Validation: 8 tests passed, 3 CUDA tests skipped; notebook format, relative links, local setup paths, Compose configuration, Linux dependency resolution, and diff checks passed. Signed-off-by: Johnny <johnnynuca14@gmail.com>
johnnynunez
changed the title
johnnynunez
marked this pull request as ready for review
October 2, 2026 19:34
…bots. Add an optional two-cube receiving-box task to the CPU and GPU lessons, with measured grasp, lift, carry, release and settling checks for SO-101 and reBot. Keep the default stack exercises and CUDA benchmark guards intact. Add opt-in notebook cells, regression tests and updated validation instructions. Signed-off-by: Johnny <johnnync13@gmail.com>
johnnynunez
changed the title
jnbntz
self-requested a review
October 5, 2026 21:17
jnbntz
approved these changes
Oct 5, 2026
johnnynunez
added a commit
to johnnynunez/accelerated-computing-hub
that referenced
this pull request
Oct 10, 2026
Refresh the MuJoCo/MJWarp lessons with the physically validated two-cube receiving-box task, matched CPU/GPU benchmark, finite viewer lifecycle, reproducible environments and unchanged raw evidence. Follow-up to NVIDIA#279. Signed-off-by: Johnny <johnnync13@gmail.com>
jnbntz
pushed a commit
that referenced
this pull request
Oct 14, 2026
Refresh the MuJoCo/MJWarp lessons with the physically validated two-cube receiving-box task, matched CPU/GPU benchmark, finite viewer lifecycle, reproducible environments and unchanged raw evidence. Follow-up to #279. Signed-off-by: Johnny <johnnync13@gmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Add
tutorials/sim2real-blogs/with three companion lessons for How to Use NVIDIA Warp and MJWarp: MuJoCo fundamentals, CPU pick and place, and batched MuJoCo Warp physics.The SO-101 and Seeed reBot DevArm can run the original stacking task or an optional receiving-box task. With
--task box, each robot grasps, lifts, carries and releases both cubes into the box. Measured checks require opposing loaded jaw contacts, airborne transport, whole-cube containment, detached settling and gripper withdrawal. The notebook cells explicitly select reference or student implementations. The default stack exercise, capacity checks and synchronized fixed-control benchmarks are preserved.The tutorial includes completed solutions, pinned MuJoCo 3.12 / Warp 1.17 dependencies, asset attribution, a dedicated Docker/Brev environment and a test entrypoint. Robot meshes download from pinned upstream revisions. The existing
tutorials/warp/tree matches the PR base.Validation
See the validation record for versions, capacities, settling criteria and tested scope. The demonstrations disclose actuator overrides and precision differences; benchmark timings measure physics stepping, not completed robot tasks or training speed.
Live Brev/Compose deployment, interactive viewers and Colab were not tested. Colab links target upstream
mainand become available after merge. Fork CI still requires the maintainers' normal review process.