不完美成就精准：将不完美数据升级用于高精度机器人操作
Imperfection for Precision: Upcycling Imperfect Data for High-Precision Robotic Manipulation
浏览论文内容
中文总结 AI 辅助
本文提出ε4P方法，通过将低精度目标任务数据和高精度不匹配任务数据在流匹配轨迹的不同噪声阶段分别利用，实现高精度机器人操作，性能提升达31.7个百分点，且可替代等量高质量数据，性能仅降4.2个百分点。
中文摘要 AI 辅助
训练用于高精度操作的视觉-语言-动作（VLA）模型通常需要特定任务的高质量数据（例如遥操作），这类数据收集起来既缓慢又昂贵。为了在不牺牲操作精度的前提下减轻这一负担，我们提出了ε4P（不完美成就精准），一种简单而有效的方法，它“升级”了两种原本会被丢弃的数据源：（1）来自目标任务的低精度数据和（2）来自不匹配任务的高精度数据。ε4P并非在协同训练中简单混合这些不完美数据源，而是控制每个数据源在流匹配轨迹中的贡献位置。具体而言，低精度的目标任务数据在高噪声阶段使用，以保留高层任务上下文；高精度的任务不匹配数据在低噪声阶段使用，以传递低层动作精度。通过在亚毫米级高精度任务和粗粒度任务上的真实机器人实验，我们证明了所提出的方法（1）能有效利用额外的不完美数据，将策略性能提升高达31.7个百分点；（2）可以用等量的不完美数据替代特定任务的高质量数据，平均性能仅下降4.2个百分点。总体而言，ε4P指向了一种可扩展的高精度操作范式，其中异构的不完美数据可以被系统地重新利用，以减少对昂贵的特定任务高质量数据的依赖。更多细节可访问此https URL。
英文摘要
Training vision-language-action (VLA) models for high-precision manipulation typically requires task-specific, high-quality data (e.g., teleoperation), which is slow and expensive to collect. To reduce this burden without compromising manipulation precision, we propose $\varepsilon$4P (Imperfection for Precision), a simple yet effective method that "upcycles" two otherwise discarded data sources: (1) low-precision data from the target task and (2) high-precision data from mismatched tasks. Rather than naively mixing these imperfect data sources throughout co-training, $\varepsilon$4P controls where each source contributes along the flow-matching trajectory. Specifically, low-precision, target-task data is used at high noise to preserve high-level task context and high-precision, task-mismatched data is used at low noise to transfer low-level action precision. Through real-robot experiments on both sub-millimeter, high-precision tasks and coarse-grained tasks, we demonstrate that the proposed method (1) effectively leverages additional imperfect data to improve policy performance by up to 31.7 percentage points, and (2) can replace an equal amount of task-specific, high-quality data with an average performance drop of only 4.2 percentage points. Overall, $\varepsilon$4P points toward a scalable paradigm for high-precision manipulation, in which heterogeneous, imperfect data can be systematically repurposed to reduce reliance on costly task-specific, high-quality data. More details are available at https://varepsilon4p.github.io/.
发表机构
- Samsung Robotics eXperience（三星机器人体验中心）
- National University of Singapore（新加坡国立大学）
- University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
机构由 AI 辅助整理，请以论文原文为准。
