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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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 previous carry controller stayed at reduced requested movement after a settled grip, while supported failed acquisitions repeated the same pinch. v0.4 gradually restores requested movement using stable contact history and can decline an unchanged retry when its evidence is clear. Native mechanics, acquisition/hold thresholds and task scoring stay fixed.
Fresh frozen comparison (seed 941827): all 84 attempts valid; each placement controller completes 4/12; all 48 open-jaw/hold controls fail. The two completed tasks that use recovery finish 2.440 and 2.640 simulated seconds sooner; four low-friction failed grips refuse 2.88–2.90 seconds earlier. Two newly perturbed heavy-object cases time out before resuming in every mode. Faster refusal is not placement success, and the unchanged settling hold is not shortened.
Complete raw attempts, source hashes, paired controls, every computation timing sample and earlier development failures are retained. The installed frozen wheel ran with network access denied, on CPU without a GPU. This is authored privileged-state simulation, not a learned policy or physical robot qualification.
Validation: 103 simulation/controller/benchmark tests, 24 Space preparation tests and 22 publisher tests passed; all 12 historical v0.3 trajectories reproduced exactly. The fresh 84-attempt receipt passed the independent consistency inspector.
Stacked on #13. Publication: research paper, Hugging Face Space, immutable Hugging Face evidence. All 293 Space payload files were verified at that revision; desktop/mobile checks passed. Verified publication receipt and pinned reproduction instructions are committed. Hosted CI status is shown separately in the PR checks.