通过可复用纠错机制，大幅降低机器人技能训练的人工干预成本。
Zero2Skill: Bootstrapping Robot Skills through Autonomous Data Collection, Training, and Deployment
- 构建可积累纠错记忆的自主数据收集系统，失败修正跨轮次复用。
- 单次尝试成功率从12.5%提升至47.5%，人工耗时降至16%。
- 适合希望降低真实机器人训练人力投入的研究者与工程师。
自主数据收集决定了操作策略学习中真实轨迹的数量与质量。现有流程通过自恢复、视觉语言模型验证或语言引导修正减少人工干预，但每次失败需重复处理，导致监督成本随会话时长线性增长而非故障种类数。本文提出Zero2Skill，一种人机协同的智能体系统，将修正措施持久化并跨轮次复用。数据收集循环实现自主采集、验证与重置，仅在重试预算耗尽时才暂停请求远程操作员介入。大语言模型解析自然语言指令，生成结构化调整存入纠错记忆，相同条件下多数失败无需再次修正。在真实桌面清理测试平台上，零2技能达成与遥操作相当的成功率，同时人工工作时间压缩至16%。语言修正使验证器-人类一致性在四个评估场景中均提升，单次尝试成功率由12.5%增至47.5%（臂选择任务：20.0%→50.0%）。基于零2技能数据微调的策略，在极低人工成本下达到遥操作训练策略的性能水平。
原文摘要 · Abstract (English)
Autonomous data collection governs the volume and quality of real-world trajectories for manipulation policy learning. Existing pipelines reduce human effort via self-resetting, VLM verification, or language-guided correction, yet episode-scoped fixes must be reissued whenever the same failure recurs, so oversight cost grows with session length rather than with the number of distinct problems. We present Zero2Skill, a human-robot symbiotic agentic system in which corrections are retained and reused across rounds. The collection loop collects, verifies, and resets autonomously, pausing for a remote operator only when a phase exhausts an explicit retry budget. An LLM parser maps each natural-language utterance to a structured adjustment stored in Corrective Memory, so addressed failure modes typically need not be corrected again under the same conditions. On a real-robot desktop-clearing testbed, Zero2Skill matches teleoperation episode success while reducing human working time to 16%. Language corrections improve verifier-human agreement in all four evaluated settings and raise average single-attempt success from 12.5% to 47.5% (arm-selection: 20.0% to 50.0%). Policies fine-tuned on Zero2Skill data match teleoperation-trained policy success at a fraction of collection human cost.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。