# 几何变化下机器人插入的视觉 Sim-to-Real 学习：在钢筋安装中的应用

> 原文链接：https://robotics.hellobike.com/papers/2026-09-17-2609-20477
> 频道：papers | 更新日期：2026-09-17 | 标签：arxiv, cs.RO
> 核心摘要：针对几何变化下钢筋插入需反复采集真机数据的问题，RebarSim全仿真训练并在真机零样本验证。

# 几何变化下机器人插入的视觉 Sim-to-Real 学习：在钢筋安装中的应用

::: v-pre
> 原文链接：[Visual Sim-to-Real Learning for Robotic Insertion under Geometric Variations: Application to Rebar Installation](https://arxiv.org/abs/2609.20477)（arXiv 2609.20477 · 发布日期：2026-09-17）

**Visual Sim-to-Real Learning for Robotic Insertion under Geometric Variations: Application to Rebar Installation**

针对几何变化下钢筋插入需反复采集真机数据的问题，RebarSim全仿真训练并在真机零样本验证。

## 核心要点

- 钢筋插入间隙仅1.4毫米，设计与制造公差变化使真机数据需随批次重采。
- 先用强化学习训练特权状态教师，再蒸馏为多视角RGB与本体输入的学生策略。
- 学生策略零样本迁移到真实世界，在真实工厂钢筋的真机rollout中成功率91.3%。
- 几何多样性、预训练、外观随机化和DAgger混合均被报告为关键；代码和资产已开放。

**验证范围与局限**：验证范围为钢筋安装插入任务，真机实验使用来自真实工厂生产批次的钢筋。

## 英文摘要 (Abstract)

Rebar insertion is among the most repetitive and physically demanding tasks on construction sites, and a contact-rich problem at 1.4 mm clearance. The parts, however, vary at two levels: a nominal design per structural member, and fabrication tolerance around each nominal design. Real-world data therefore has to be re-collected as designs and batches change. We present RebarSim, a visual sim-to-real system trained entirely in simulation. A privileged state-based teacher is trained with reinforcement learning over procedurally generated rebar geometries, then distilled into a multi-view student that maps raw RGB and proprioception directly to actions under extensive domain randomization. The student transfers to the real world zero-shot, seating rebars taken from a real factory production run in 91.3% of real-robot rollouts. Underlying that result, geometry diversity and pretraining both bring benefits. Training across a diverse set of nominal designs rather than one lifts the zero-shot success of both the teacher and the student on unseen designs, and the student policy outperforms a single-design specialist on that specialist's own design. A pretrained student then adapts to a new design with 4--6x fewer distillation samples than one trained from scratch. Visual sim-to-real transfer depends on appearance randomization and the DAgger mixture: removing either one sharply lowers success. Videos, code, and task assets are available at https://rebarsim.github.io.

作者：Tao Sun, Beining Han, Patrick Yin, Rui Xu, Harry He, Abhishek Gupta, Szymon M. Rusinkiewicz, Yi Shao
:::

## 原始参考来源 (Sources)

- [arXiv 2609.20477：Visual Sim-to-Real Learning for Robotic Insertion under Geometric Variations: Application to Rebar Installation](https://arxiv.org/abs/2609.20477)