发表机构
Carnegie Mellon University（卡内基梅隆大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
ARSTAG是一个智能体式Real2Sim2Real系统，从单张RGB图像和自然语言指令自动生成机器人策略学习数据，通过语言智能体层级构建场景、生成演示并随机化扩展，在七个操作任务中实现三种策略架构的仿真到现实迁移，pi0.5达到74.6%的真实世界成功率。
AI 中文摘要
将视觉运动策略适应于新的操作任务通常需要大量的人工工程或远程操作数据收集。仿真可以在规模上提供任务特定的数据，但构建场景、设计专家行为以及配置数据生成仍然需要大量的每任务工作量。我们提出了ARSTAG，一个智能体式Real2Sim2Real系统，它直接将单个RGB图像和自然语言指令转化为机器人策略学习数据。一个语言智能体层级结构构建任务范围的仿真场景，生成机器人可行的演示，并通过任务一致的随机化扩展训练分布，同时一个协调智能体管理跨阶段的反馈和恢复。在涵盖抓取、放置和堆叠的七个操作任务中，ARSTAG生成的演示使三种视觉运动策略架构能够实现到双臂机器人的仿真到现实迁移，其中pi0.5实现了74.6%的平均真实世界成功率。消融研究表明，任务一致的随机化显著提高了鲁棒性，且策略性能随生成数据集规模的增大而提升。项目网页：此https URL。
英文摘要
Adapting visuomotor policies to new manipulation tasks often requires substantial manual engineering or teleoperated data collection. Simulation can provide task-specific data at scale, but constructing the scene, designing expert behavior, and configuring data generation still require significant per-task effort. We present ARSTAG, an agentic Real2Sim2Real system that turns a single RGB image and a natural-language instruction directly into robot policy-learning data. A hierarchy of language agents constructs a task-scoped simulation scene, generates robot-feasible demonstrations, and expands the training distribution through task-consistent randomization, while a coordinator agent manages cross-stage feedback and recovery. Across seven manipulation tasks spanning grasping, placement, and stacking, the ARSTAG-generated demonstrations enable sim-to-real transfer of three visuomotor policy architectures to a dual-arm robot, with pi0.5 achieving an average real-world success rate of 74.6%. Ablations show that task-consistent randomization substantially improves robustness, and policy performance increases with generated dataset size. Project webpage: https://boweili666.github.io/ARSTAG/.
Comments12 pages, 15 figures and tables. Project webpage: https://boweili666.github.io/ARSTAG/
