发表机构
Nanyang Technological University; Beihang University; National University of Singapore; Beijing Aerospace Control Instrument Research Institute（南洋理工大学; 北京航空航天大学; 新加坡国立大学; 北京航天测控仪器研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出轻量、模型无关的 PhyFilter 反馈机制，无需海量训练数据，即可提升机器人在动态不确定性下的泛化能力，在四类机器人系统上验证了其有效性。
AI 中文摘要
生物通过自身固有物理结构和终身反馈驱动学习，展现出对未知环境的极强适应性。赋予机器人类似的泛化能力对其在真实世界的可靠运行至关重要。尽管近期有研究尝试通过扩大训练数据规模提升泛化性，但对于机器人领域而言，该策略并不实用——收集类似大语言模型规模的真实世界演示数据成本极高且速度极慢。与依赖海量数据的思路相反，本文表明机器人在动态不确定性下，即便使用有限训练数据，也可通过名为 PhyFilter 的反馈机制实现有效泛化；该机制通过物理滤波后的学习残差修正学习输出，作为轻量、模型无关的模块，其参数可通过自动学习算法自动优化，无需手动调参，可无缝集成至各类机器人策略中。我们在四个典型机器人系统上验证了 PhyFilter：四足机器人可泛化至未知地形、负载变化和速度范围；无人机可在未知风扰下飞行；空中机械臂可在风扰和质量不确定性下实现厘米级空中抓取；加速度微分器可在分布偏移下保持鲁棒性。这些结果表明，物理滤波反馈可作为海量数据扩展的有力替代方案。
英文摘要
Living organisms exhibit extraordinary adaptability to unseen environments through their intrinsic physical structures and lifelong feedback-driven learning. Endowing robots with comparable generalization is critical for reliable operation in the real world. While recent approaches attempt to improve generalization by scaling training data, such strategies remain impractical for robotics, where collecting real-world demonstrations at the scale of large language models is prohibitively costly and slow. Contrary to this reliance on massive datasets, we show that robots can generalize effectively under dynamics uncertainties even with limited training data by leveraging a feedback mechanism, namely PhyFilter, that corrects learning outputs with physics-filtered learning residuals. PhyFilter operates as a lightweight, model-agnostic module whose parameters can be automatically optimized through an auto-learning algorithm, eliminating manual tuning and enabling seamless integration with diverse robot policies. We validate PhyFilter across four representative robotic systems, demonstrating that it enables quadruped robots to generalize to unseen terrains, payload variations, and speed ranges; drones to flight under unseen wind disturbances; aerial manipulators to achieve centimeter-level in-air capture despite wind and mass uncertainties; and acceleration differentiators to remain robust with distribution shift. These results show that physics-filtered feedback can serve as a powerful alternative to massive data scaling.
CommentsAccepted by npj Robotics
