外观
论文arXiv 2610.1046221 天前
FoldBack：用于长程服装折叠的自校正掩码生成策略
原文链接：FoldBack: Self-Correcting Masked Generative Policy for Long-Horizon Garment Folding（arXiv 2610.10462 · 发布日期：2026-10-07）
FoldBack: Self-Correcting Masked Generative Policy for Long-Horizon Garment Folding
FoldBack面向长程服装折叠中的失误延续问题，以推理时自校正掩码生成策略在真实服装上验证。
核心要点
- 现有长轨迹策略在漏抓或滑抓后，可能在衣物未达预期形态时继续执行。
- FoldBack围绕何时细化验证、如何回滚、在哪里及如何重试三类推理时决策设计恢复。
- 它回到可重试的预抓取配置，保留成功抓取，并重生成失败片段和部分未来动作。
- 在33件六类真实服装上，达75.2%成功率和0.837 final-mask IoU，高于最强既有基线。
验证范围与局限：验证范围为33件、六类真实服装的长程服装折叠任务。
英文摘要 (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.
作者：Lipeng Zhuang, Shiyu Fan, Yingdong Ru, Zhuo He, Florent P. Audonnet, Paul Henderson, Gerardo Aragón-Camarasa
来源
本页事实对应的公开页面，以原文为准。