学习折叠技巧：在2026年LeHome挑战赛中成为获奖解决方案（线上第一名，线下第二名）
Learning to Fold: prizewinning solution at LeHome Challenge 2026 (1st place online, 2nd offline)
摘要
我介绍了自己针对LeHome Challenge 2026比赛的解决方案。这是一场由ICRA 2026组织举办的双手服装折叠竞赛。该系统在线上模拟阶段中获得了62个团队中的第1名，在现实世界决赛中则获得了第2名。该系统通过强化学习机制来优化视觉-语言-动作策略。该策略本身具有价值函数：用于预测动作的同一网络也用于预测成功程度、进展情况以及与任务相关的未来指标。这些预测结果被用于评估优势、实时检测失败情况以及选择合适的候选方案。这项研究主要将现有的强化学习理念与工程技术和优化方法相结合，这些方法可以单独使用，也可以一起使用：AWR与RECAP技术结合用于流匹配型视觉-语言-动作策略；通过HuggingFace Hub实现异步分布式训练/部署流程；通过Thompson采样技术优化推理阶段的超参数；采用相机对齐工具、大量增强技术以及类似DAgger的HIL数据收集方法来实现从仿真到实时的转换。
English Abstract
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.