# MAIA：通过多模态世界模型预判接触丰富操作中的交互异常

> 原文链接：https://robotics.hellobike.com/papers/2026-10-15-2610-21686
> 频道：papers | 更新日期：2026-10-15 | 标签：arxiv, cs.RO
> 核心摘要：MAIA用视觉-动作-力世界模型预测计划动作后果并提前预判接触操作异常，在真实多模态数据集和新旧任务上验证。

# MAIA：通过多模态世界模型预判接触丰富操作中的交互异常

::: v-pre
> 原文链接：[MAIA: Anticipating Interaction Anomalies in Contact-Rich Manipulation via Multimodal World Models](https://arxiv.org/abs/2610.21686)（arXiv 2610.21686 · 发布日期：2026-10-15）

**MAIA: Anticipating Interaction Anomalies in Contact-Rich Manipulation via Multimodal World Models**

MAIA用视觉-动作-力世界模型预测计划动作后果并提前预判接触操作异常，在真实多模态数据集和新旧任务上验证。

## 核心要点

- 接触丰富操作需要及早发现执行偏差，但多数监测方法只能在异常可观测后反应。
- MAIA学习正常交互动力学，预测计划动作的视觉与力响应，并用轻量异常头预判即将发生的异常。
- 论文引入CRIA真实多模态数据集，含2,852条轨迹、18个任务及帧级异常标注。
- 在已见和新任务上，MAIA较代表性基线异常判别更准确且预警更早，并提升异常感知动作选择的任务成功率。

**验证范围与局限**：验证范围为真实世界CRIA数据集中的18个接触丰富操作任务，以及已见和新任务实验。

## 英文摘要 (Abstract)

Reliable contact-rich manipulation requires detecting execution deviations early enough for corrective action. Yet most monitoring methods are reactive, identifying anomalies only after they become observable. Anticipating such anomalies is challenging because visual observations only partially reveal physical interaction, while planned actions describe intended motion rather than their actual physical consequences. We address this problem with MAIA, a multimodal framework for anomaly anticipation. Its vision-action-force world model learns nominal interaction dynamics and predicts the visual and force responses of planned actions, while a lightweight anomaly head predicts impending interaction anomalies. To enable systematic evaluation, we introduce CRIA, a real-world multimodal dataset of 2,852 trajectories spanning 18 contact-rich manipulation tasks, with frame-level annotations of anomaly onset and evolution. Experiments on both seen and novel tasks show that MAIA achieves more accurate anomaly discrimination than representative baselines while providing earlier warnings, with bACC scores of 0.765 and 0.710 and median anticipation lead times of 28.0 and 25.1 steps, respectively. Moreover, using these predictions for anomaly-aware action selection improves end-to-end task success from 52.6% to 61.5% without retraining the underlying manipulation policy. These results highlight the value of modeling future physical consequences for proactive anomaly anticipation.

作者：Yi Qiao, Maolin Zheng, Haoran Sun, Lujia Dong, Wenzhao Lian
:::

## 原始参考来源 (Sources)

- [arXiv 2610.21686：MAIA: Anticipating Interaction Anomalies in Contact-Rich Manipulation via Multimodal World Models](https://arxiv.org/abs/2610.21686)