FoldBack improves long-horizon garment folding with self-correcting robot policy
FoldBack: Self-Correcting Masked Generative Policy for Long-Horizon Garment Folding
RoboticsArtificial Intelligence
Summary
Folding clothes with robots is hard because if the robot misses a grab or slips, it often just continues without fixing the problem. The authors created FoldBack, a system that helps the robot notice mistakes, go back to fix them, and then try again without having to retrain or get special examples of recoveries. This system decides when to check if the fold is right, undoes wrong moves, and avoids repeated failures. They tested it on many types of real clothes and showed it folds better than previous systems.
What this means in practice
- •For industrial robotics teams: Enhance robot garment folding lines to detect and fix grasp failures without retraining, improving folding success and efficiency.
- •For home robotics developers: Develop more reliable home robots able to fold clothes by self-correcting mistakes during long folding sequences.
Authors
Lipeng Zhuang, Shiyu Fan, Yingdong Ru, Zhuo He, Florent P. Audonnet, Paul Henderson, Gerardo Aragon Camarasa
Abstract
We present FoldBack, a self-correcting masked generative policy for long-horizon garment folding. Existing long-trajectory policies may continue after a missed or slipped grasp even when the garment has not reached the intended configuration. We structure FoldBack's recovery mechanisms around three inference-time decisions: when to refine and verify, how to roll back, and where and how to retry. FoldBack aligns refinement and grasp verification with pick-and-place events, returns the robot to a retryable pre-grasp configuration while preserving successful grasps, and selectively regenerates the failed segment and selected future actions while avoiding previous failed grasp locations. To our knowledge, FoldBack is the first editable full-trajectory policy to unify these decisions, enabling failed interactions to be detected, undone, and repaired before execution continues, without recovery demonstrations or base-policy retraining. Across 33 real garments from six categories, FoldBack achieves 75.2% final folding success and 0.837 final-mask IoU, versus 45.7% and 0.689 for the strongest prior baseline.