arXiv 2604.113512026-04-13cs.RO

WM-DAgger: 通过世界模型实现模仿学习中的高效数据聚合

WM-DAgger: Enabling Efficient Data Aggregation for Imitation Learning with World Models

Peking University（北京大学）；JD Logistics（京东物流）；Nankai University（南开大学）；The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Rutgers University（罗格斯大学）；University of Texas at Dallas（德克萨斯大学达拉斯分校）；Institut Polytechnique de Paris（巴黎综合理工学院）

Anlan Yu, Zaishu Chen, Peili Song, Zhiqing Hong, Haotian Wang, Desheng Zhang, Tian He, Yi Ding, Daqing Zhang

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出WM-DAgger框架，利用世界模型合成OOD恢复数据，无需人工干预，提升机器人模仿学习效率，实验表明在少样本下显著提高任务成功率。

摘要：

模仿学习是训练机器人策略的强大范式，但其性能受限于累积误差：微小策略不准确可能导致机器人进入训练集中未见过的分布外（OOD）状态，从而生成更大误差并最终失败。尽管数据聚合（DAgger）框架试图解决此问题，但其依赖持续的人工参与严重限制了可扩展性。本文提出WM-DAgger，一种高效的数据显示聚合框架，利用世界模型合成OOD恢复数据而无需人工参与。具体而言，我们专注于具有眼在手机械臂的操作任务，并仅使用少量示例。为避免合成误导数据并克服世界模型固有的幻觉问题，我们的框架引入两个关键机制：（1）纠正行动合成模块，生成任务导向的恢复动作以防止误导监督；（2）一致性引导过滤模块，通过将终端合成帧锚定到专家示范中的相应真实帧，剔除物理上不可能的轨迹。我们广泛验证了WM-DAgger在多个现实世界机器人任务上的性能。结果表明，我们的方法显著提高了成功率，在软袋推动任务中仅使用五个示例就达到了93.3%的成功率。源代码可在https://github.com/czs12354-xxdbd/WM-Dagger公开获取。

英文摘要：

Imitation learning is a powerful paradigm for training robotic policies, yet its performance is limited by compounding errors: minor policy inaccuracies could drive robots into unseen out-of-distribution (OOD) states in the training set, where the policy could generate even bigger errors, leading to eventual failures. While the Data Aggregation (DAgger) framework tries to address this issue, its reliance on continuous human involvement severely limits scalability. In this paper, we propose WM-DAgger, an efficient data aggregation framework that leverages World Models to synthesize OOD recovery data without requiring human involvement. Specifically, we focus on manipulation tasks with an eye-in-hand robotic arm and only few-shot demonstrations. To avoid synthesizing misleading data and overcome the hallucination issues inherent to World Models, our framework introduces two key mechanisms: (1) a Corrective Action Synthesis Module that generates task-oriented recovery actions to prevent misleading supervision, and (2) a Consistency-Guided Filtering Module that discards physically implausible trajectories by anchoring terminal synthesized frames to corresponding real frames in expert demonstrations. We extensively validate WM-DAgger on multiple real-world robotic tasks. Results that our method significantly improves success rates, achieving a 93.3\% success rate in soft bag pushing with only five demonstrations. The source code is publicly available at https://github.com/czs12354-xxdbd/WM-Dagger.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。