# PredTac：利用预测触觉学习富接触操作

> 原文链接：https://robotics.hellobike.com/papers/2026-09-14-2609-15198
> 频道：papers | 更新日期：2026-09-14 | 标签：arxiv, cs.RO
> 核心摘要：PredTac从因果视觉与机器人状态预测触觉，用作策略接口，并在仿真和真机三项富接触任务中验证。

# PredTac：利用预测触觉学习富接触操作

::: v-pre
> 原文链接：[PredTac: Learning Contact-Rich Manipulation with Predicted Touch](https://arxiv.org/abs/2609.15198)（arXiv 2609.15198 · 发布日期：2026-09-14）

**PredTac: Learning Contact-Rich Manipulation with Predicted Touch**

PredTac从因果视觉与机器人状态预测触觉，用作策略接口，并在仿真和真机三项富接触任务中验证。

## 核心要点

- 真实触觉传感器带来硬件、标定、同步与维护成本，增加学习和部署复杂度。
- 先用触觉监督训练预测器，再在下游训练和执行中以预测触觉提供接触信息。
- 仿真目标偏移评估中，三任务成功率为27.0%、52.0%、44.7%，优于视觉基线。
- 真机上预测触觉ACT三任务均值70.0%，接近测量触觉72.2%，高于视觉21.1%。

**验证范围与局限**：验证范围为仿真和一台真机上的三项富接触操作任务。

## 英文摘要 (Abstract)

Contact-rich manipulation benefits from tactile feedback, yet physical tactile sensors introduce hardware, calibration, synchronization, and maintenance costs that complicate policy learning and deployment. We formulate predicted touch as an alternative to measured tactile input and present PredTac, a framework that learns to infer tactile states from causal visual observations and robot states and uses the predicted touch as an explicit interface for policy learning and execution. A tactile predictor is first trained with tactile supervision and then used to provide contact information without requiring measured tactile input during downstream policy training or execution. We evaluate PredTac across three contact-rich manipulation tasks in simulation and on a real robot, and further examine how policy performance depends on the predicted contact content. In simulation goal-offset evaluations, predicted-touch policies achieve 27.0%, 52.0%, and 44.7% success on USB, Barbed-spike, and Valve, respectively, improving over the visual baseline by 8.0-13.7 percentage points. On the real robot, predicted-touch ACT achieves 70.0%, 50.0%, and 90.0% success on USB insertion, Barbed extraction, and Valve rotation, respectively, with a three-task mean of 70.0%, approaching measured-touch ACT at 72.2% and substantially outperforming visual ACT at 21.1%. Fixed-policy interventions further show that performance is sensitive to the spatial structure of predicted contact, with spatial rearrangement at fixed value distributions reducing Valve success by 10.7 percentage points. These results demonstrate that predicted touch can provide useful contact information for contact-rich manipulation without requiring tactile sensing as a policy input.

作者：Weijia Fan, Daqiang Guo
:::

## 原始参考来源 (Sources)

- [arXiv 2609.15198：PredTac: Learning Contact-Rich Manipulation with Predicted Touch](https://arxiv.org/abs/2609.15198)