从视频学习可变形关节物体的弹塑性动态，实现长期精准预测。
BoxTwin: Learning Elastoplastic Articulated Object Dynamics from Videos
- 基于视频重建与物理感知本构模型，捕捉非线性弹性与塑性变形
- 在手动折叠和双臂操作中，长期追踪关节轨迹并复现接触后塑性行为
- 适合需要高精度物理交互预测的机器人操控场景
数字孪生使机器人能够预判和适应物理交互，但现有模型难以处理具有非线性弹性、塑性屈服和损伤累积特性的可变形关节物体（EAOs）。我们提出 BoxTwin，一个从视频中学习 EAO 全动态的交互式数字孪生框架。该流程重建场景，并为每个 EAO 识别出物理感知的本构模型。在人工折叠和双臂操作 EAOs 的实验中，BoxTwin 能够准确追踪关节轨迹，并在长时间序列上重现接触后的塑性行为。通过结合视频驱动重建与弹塑性损伤建模，BoxTwin 推动数字孪生向非结构化环境中可变形关节物体的预测性、自适应控制迈进。
原文摘要 · Abstract (English)
Digital twins enable robots to anticipate and adapt to physical interactions, but existing models struggle with elastoplastic articulated objects (EAOs) that exhibit nonlinear elasticity, plastic yielding, and damage accumulation. We present BoxTwin, an interactive digital twin framework that learns the full dynamics of EAOs from videos. Our pipeline reconstructs the scene, identifies a physics aware constitutive model for each EAO. Experiments on manual folding and dual arm manipulation of EAOs show that BoxTwin accurately tracks joint trajectories and reproduces post contact plastic behavior over long horizons. By integrating video driven reconstruction with elastoplastic damage modeling, BoxTwin advances digital twins toward predictive, adaptive control of deformable articulated objects in unstructured environments.
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。