让机器人动态模型快速适应新环境，实现精准预测控制。
Adapting Neural Robot Dynamics on the Fly for Predictive Control
- 离线训练神经模型，线上低秩更新参数
- 实测在新环境下仍保持稳定追踪控制
- 适合需要快速响应的实时机器人系统
精确的动态模型对自主移动机器人的预测控制设计至关重要。基于物理的模型通常过于简化，无法捕捉真实世界效应；而数据驱动模型则依赖大量数据且训练缓慢。本文提出一种快速适应神经机器人动态模型的方法，结合离线训练与高效在线更新。该方法离线学习增量式神经动态模型，并在线执行低秩二阶参数自适应，实现无需全量重训的快速更新。我们在真实四旋翼无人机上验证了该方法，在新型工况下实现了鲁棒的预测跟踪控制。
原文摘要 · Abstract (English)
Accurate dynamics models are critical for the design of predictive controller for autonomous mobile robots. Physics-based models are often too simple to capture relevant real-world effects, while data-driven models are data-intensive and slow to train. We introduce an approach for fast adaptation of neural robot dynamic models that combines offline training with efficient online updates. Our approach learns an incremental neural dynamics model offline and performs low-rank second-order parameter adaptation online, enabling rapid updates without full retraining. We demonstrate the approach on a real quadrotor robot, achieving robust predictive tracking control in novel operational conditions.
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。