[论文] SIM1: Physics-Aligned Simulator as Zero-Shot Data Scaler in Deformable...

## 论文概要 **研究领域**: CV **作者**: Yunsong Zhou, Hangxu Liu, Xuekun Jiang **发布时间**: 2025-04-10 **arXiv**: [2504.07903](https://arxiv.org/abs/2504.07903) ## 中文摘要 可变形物体的机器人操作代表了具身学习中的数据密集型领域，其中形状、接触和拓扑以远超刚体变化的方式共同演化。虽然仿真有望缓解真实世界数据获取的成本，但主流的仿真到现实流程仍根植于刚体抽象，产生不匹配的几何形状、脆弱的软体动力学以及不适合布料交互的运动基元。我们认为仿真的失败不在于其合成性质，而在于其缺乏物理基础。为解决这一问题，我们提出了SIM1，一个物理对齐的从现实到仿真再到现实的数据引擎，将仿真锚定在物理世界中。给定有限的演示，系统将场景数字化为度量一致的双胞胎，通过弹性建模校准可变形动力学，并通过基于扩散的轨迹生成与质量过滤来扩展行为。这一流程将稀疏观测转换为具有接近演示保真度的规模化合成监督。实验表明，仅在合成数据上训练的策略在1:15的等效比例下达到与真实数据基线相当的性能，同时在真实世界部署中实现90%的零样本成功率和50%的泛化增益。这些结果验证了物理对齐仿真作为可变形操作的可扩展监督和高效数据策略学习的实用路径。 ## 原文摘要 Robotic manipulation with deformable objects represents a data-intensive regime in embodied learning, where shape, contact, and topology co-evolve in ways that far exceed the variability of rigids. Although simulation promises relief from the cost of real-world data获取, prevailing sim-to-real pipelines remain rooted in rigid-body abstractions, producing mismatched geometry, fragile soft dynamics, and motion primitives poorly suited for cloth interaction. We posit that simulation fails not for being synthetic, but for being ungrounded. To address this, we introduce SIM1, a physics-aligned real-to-sim-to-real data engine that grounds simulation in the physical world. Given limited demonstrations, the system digitizes scenes into metric-consistent twins, calibrates deformable dynamics through ... --- *自动采集于 2026-04-12* #论文 #arXiv #CV #小凯