Repository navigation
Conversation
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The existing authored arm has simplified links and disabled moving-arm collision masks, so it cannot establish hardware fidelity. This adds a separate pinned UR5e/Robotiq reference with verified in-memory assets, coupled gripper mechanics, collision inventory, three-timestep CPU diagnostics and offline replay. A strict paired trace scorer measures simulator/physical joint, tool and jaw disagreement without declaring hardware readiness.
Explicit provisioning preserves per-asset licenses; installed verification and scoring work without the optional native runtime. The measured bundle retains all 18 unloaded numerical attempts, source hashes and narrow timing boundaries. Physical calibration, shoe task learning and the bimanual assembly remain unmeasured. LeRobot can be a future collection layer while DVIDIA builds its own policy.
Validation: 488 local tests passed, including actual pinned native assets; 36 new reference/CLI tests and 25 paired-scoring tests pass. Installed-wheel help, asset verification, receipt inspection and scoring pass without NumPy/MuJoCo. Offline report controls, downloads and layouts passed at 1440, 390 and 320 pixels with no browser errors or remote requests. A new Linux CI job explicitly provisions and tests the reference assets.
This PR builds on the compact actuator pilot in #8. The new reference does not reuse that artifact as a task policy.