[robotic_grounding] enable object orientation tracking for ReconBody - #154
Open
Arkitect-z wants to merge 1 commit into
Open
Arkitect-z wants to merge 1 commit into
Arkitect-z wants to merge 1 commit into
Conversation
Signed-off-by: Arkitect-z <1458006823@qq.com>
Arkitect-z
marked this pull request as ready for review
August 24, 2026 21:42
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
motion_object_orientation_error_expprimitive inG1SonicReconBodyRewardsCfg.Rationale
ReconBody already tracks object position, while the repository's existing object-orientation primitive is not included in its reward composition. The CHORD paper describes the task objective as tracking object pose in
SE(3).This change uses two independent exponential kernels:
exp(-||delta_position||^2 / 0.2^2) + exp(-theta^2 / 0.4^2).It therefore aligns ReconBody with the paper's object-pose objective, but it is not a formula-level reproduction of the paper's single joint
SE(3)exponential kernel.Adding the proposed
weight = 1.0term raises the maximum object-pose reward contribution from1.0to2.0when both errors are zero. The paper does not specify this weight orstd = 0.4: the weight is proposed for consistency with the existing position and orientation-tracking term weights, and the standard deviation is inferred from the existing ReconBody anchor and end-effector orientation kernels. Maintainer confirmation of both values is requested.Scope
motion_object_position_error_expwith its existing function,weight = 1.0, andstd = 0.2.Test plan
robotic_grounding/**component pre-commit passed.git diff --checkpassed.python3 -m compileall -q robotic_grounding/sourcepassed.Checklist