# MIRROR：从离线机器人交互数据中反映隐藏对称性

> 原文链接：https://robotics.hellobike.com/papers/2026-10-12-2610-17422
> 频道：papers | 更新日期：2026-10-12 | 标签：arxiv, cs.RO
> 核心摘要：MIRROR从离线交互数据学习观测-动作联合对称变换，并在Robosuite任务与真机零样部署中验证。

# MIRROR：从离线机器人交互数据中反映隐藏对称性

::: v-pre
> 原文链接：[MIRROR: Reflecting Hidden Symmetries from Offline Robot Interaction Data](https://arxiv.org/abs/2610.17422)（arXiv 2610.17422 · 发布日期：2026-10-12）

**MIRROR: Reflecting Hidden Symmetries from Offline Robot Interaction Data**

MIRROR从离线交互数据学习观测-动作联合对称变换，并在Robosuite任务与真机零样部署中验证。

## 核心要点

- 像素观测下物理对称会产生视角相关复杂图像变换，难以手工指定。
- MIRROR在POMDP中学习已知对称群对观测和动作的联合作用。
- 方法用一步转移元组的对抗匹配耦合变换，并以群一致性正则约束。
- 四个Robosuite桌面任务中三项指标聚合误差最低，BC平均成功率升至85.12%。

**验证范围与局限**：验证包括四个Robosuite桌面操作任务，以及在仿真高度图和分割掩码观测上训练后零样部署到一台Kinova GEN3真机。

## 英文摘要 (Abstract)

Symmetry-aware robot policy learning requires coupled observation-action transformations. For pixel observations, even simple physical symmetries can induce complex, viewpoint-dependent image transformations that are difficult to specify by hand. Existing symmetry-aware methods often rely on representations with predefined transformations, requiring preprocessing tailored to the sensing setup and task. We introduce MIRROR, a framework that learns how a known symmetry group operates jointly on observations and actions in partially observable Markov decision processes (POMDPs) from offline interaction data. MIRROR couples these transformations through adversarial tuple matching over one-step transition tuples, while explicit group-consistency regularization penalizes violations of the group-operation axioms. On four Robosuite tabletop manipulation tasks, MIRROR achieves the lowest aggregate error on three of four symmetry-quality metrics, including a 61% reduction in compatibility error (which measures consistency under transformation composition) relative to an adapted LaLiGAN baseline. Using the learned transformations in frame-averaged behavior cloning (BC) yields higher mean success rates than vanilla BC on all four tasks, increasing average success from 80.48% to 85.12%, with per-task gains up to 9.66 percentage points. We also repeat symmetry discovery and policy training in simulation with heightmap and segmentation-mask observations, then deploy the resulting policy zero-shot on a real Kinova GEN3 arm. Project page: https://pomdp-mirror.github.io.

作者：Shiqi Liu, Yihang Yao, Peide Huang, Mengdi Xu, Ding Zhao
:::

## 原始参考来源 (Sources)

- [arXiv 2610.17422：MIRROR: Reflecting Hidden Symmetries from Offline Robot Interaction Data](https://arxiv.org/abs/2610.17422)