学习折叠：2026年LeHome挑战赛获奖方案（线上第一名，线下第二名）

摘要

我描述了针对LeHome Challenge 2026（ICRA 2026双臂衣物折叠竞赛）的解决方案。该系统在线上（仿真）赛段中于62支队伍中排名第一，并在实物决赛中位列第二。该方法通过强化学习循环改进了视觉-语言-动作（VLA）策略。该策略本身即是其价值函数：同一网络在预测动作的同时，也预测成功率、进度以及若干与任务相关的未来量，这些预测被用于优势估计、实时故障检测和候选动作筛选。本工作主要将现有强化学习思想与工程及优化贡献重新组合，这些贡献既可作为一个完整方案使用，也可单独采用：将AWR与RECAP结合用于流匹配VLA；基于HuggingFace Hub的异步分布式训练/推演管线；通过汤普森采样实现推理时超参数优化；涵盖相机对齐工具、强数据增强及类似DAgger的人类在线学习数据收集的仿真到现实方案。

English

I describe my solution to the LeHome Challenge 2026, an ICRA 2026 competition on bimanual garment folding. The system placed 1st of 62 teams in the online (simulation) round and 2nd in the real-world final. It improves a vision-language-action (VLA) policy with a reinforcement-learning loop. The policy is its own value function: the same network that predicts actions also predicts success, progress, and a few task-relevant future quantities, and those predictions drive advantage estimation, live failure detection, and candidate selection. The work mostly recombines existing RL ideas with engineering and optimization contributions that can be used together as one recipe or individually: AWR + RECAP combined for flow-matching VLA; an asynchronous distributed training / rollout pipeline through HuggingFace Hub; inference-time hyperparameters optimization via Thompson sampling; a sim-to-real recipe with camera-alignment tooling, heavy augmentation and DAgger-like HIL data collection.