# VioLA：从人类数据学习通用人形机器人控制策略

> 原文链接：https://robotics.hellobike.com/papers/2026-10-08-2610-12435
> 频道：papers | 更新日期：2026-10-08 | 标签：arxiv, cs.RO
> 核心摘要：VioLA通过预测身体与手部运动潜变量并由控制器执行，利用人类数据训练，在真机上验证零样本运动和操作。

# VioLA：从人类数据学习通用人形机器人控制策略

::: v-pre
> 原文链接：[VioLA: Learning Generalist Humanoid Control Policies from Human Data](https://arxiv.org/abs/2610.12435)（arXiv 2610.12435 · 发布日期：2026-10-08）

**VioLA: Learning Generalist Humanoid Control Policies from Human Data**

VioLA通过预测身体与手部运动潜变量并由控制器执行，利用人类数据训练，在真机上验证零样本运动和操作。

## 核心要点

- 问题在于人形机器人全身动作空间大且耦合，机器人示教稀缺，难以直接泛化到新指令。
- VioLA不预测关节命令，而预测身体和手部运动潜变量，并由预训练控制器在机器人上执行。
- 运动编码器将人类与机器人运动映射到同一潜空间，使包含140.6百万帧的数据可用于训练。
- 真机零样本运动成功率达100%，操作成功率达88.6%；代码和检查点将发布。

**验证范围与局限**：验证范围为真实机器人上的运动指令和操作任务，并跨两种VLA与一种world-action模型骨干测试。

## 英文摘要 (Abstract)

Teaching a humanoid to follow instructions with its whole body runs into two obstacles. Its action space is large and tightly coupled: legs, arms, and fingers must move together while the robot keeps its balance, which makes joint-level actions hard to learn. And humanoid demonstrations are scarce, so current humanoid generalist policies do not follow new instructions out of the box and are fine-tuned on teleoperated demonstrations of each task before deployment. Human demonstrations exist in far larger numbers, but a person's motion is not a robot command. We remove both obstacles by changing what the generalist policy predicts. We introduce VioLA, a generalist humanoid policy that predicts body and hand motion latents instead of joint commands. A pretrained body- and hand-controller execute these latents on the robot. Their corresponding motion encoders map human and robot motion into the same latent spaces. A human recording is therefore labeled in the policy's action space, and the training demonstration pool contains 140.6 million frames, 93.2% of them human. As a result, VioLA follows locomotion instructions on the real robot zero-shot, without task-specific fine-tuning, reaching 100% success where GR00T N1.7 and $Ψ_0$ reach 16.7% and 0%, respectively. It also reaches 88.6% manipulation success without task-specific fine-tuning. The same approach works across two VLA and one world-action model backbones. A generalist policy trained on human demonstrations alone performs locomotion tasks on the real robot zero-shot. Code and checkpoints will be released.

作者：Berat Mert Albaba, Jens Beißwenger, Anna Manasyan, Daniel Marta, Michael J. Black, Wieland Brendel, Andreas Krause, Georg Martius, Martin Riedmiller
:::

## 原始参考来源 (Sources)

- [arXiv 2610.12435：VioLA: Learning Generalist Humanoid Control Policies from Human Data](https://arxiv.org/abs/2610.12435)