# ROAM系统融合VLM与RL实现足式机器人公里级野外自主导航

> 原文链接：https://robotics.hellobike.com/papers/2026-11-20-2611-25215-a3a7dd4c
> 频道：papers | 更新日期：2026-11-20 | 标签：auto, arxiv
> 核心摘要：ROAM系统结合视觉语言模型常识推理与强化学习运动策略，使足式机器人在高山及城市非结构化环境完成公里级自主导航。通过开放词汇分割生成实时风险地图耦合高低层控制，验证了基础模型在复杂野外部署中的有效性与局限。

# ROAM系统融合VLM与RL实现足式机器人公里级野外自主导航

::: v-pre
> 本文由 AI 根据公开内容整理，可能有误，请以原文为准：[Bridging vision-language reasoning and learned control for legged navigation in the wild](https://arxiv.org/abs/2611.25215)（arXiv 2611.25215）

## 核心内容

针对足式机器人在复杂非结构化环境中兼顾低层运动控制与高层语义理解的难题，研究团队提出ROAM（Robust Open-world Autonomous Mobility）系统。该系统成功支持足式机器人在高山与城市混合场景中实现公里级自主徒步，具备识别警示牌、评估地形风险及最小人工干预能力，标志着具身智能在野外可靠部署取得阶段性进展，同时揭示了当前VLM在空间推理方面的现存瓶颈。

技术架构上，ROAM利用互联网规模预训练的VLM提供场景语义解释与风险评估，并通过开放词汇分割模块将标签转化为稠密实时风险代价图。该代价图作为条件输入，驱动仿真中预训练的强化学习敏捷运动与局部规划策略，实现了高层语义推理与底层物理控制的紧耦合。实验表明该机制能有效引导机器人避开危险区域并选择安全落脚点，但系统性能仍受限于基础模型本身的可靠性与空间感知精度。

## 关键看点

- **系统架构**：融合VLM常识推理与仿真RL策略，通过开放词汇分割生成实时风险代价图，实现语义理解与底层运动控制的紧耦合
- **实测性能**：在高山与城市复杂非结构化环境中完成公里级自主导航，具备标识识别与地形风险评估能力，人工干预极少
- **技术局限**：验证了基础模型是野外导航关键组件但非完备解，当前VLM在空间推理精度与长期可靠性方面仍存在显著挑战

## 工程与产业观察

ROAM验证了“VLM语义代价图+RL底层控制”范式在公里级野外导航的工程可行性，为非结构化环境具身部署提供了参考架构。但其暴露的VLM空间推理短板提示，单纯依赖基础模型无法解决所有长尾问题，未来需重点攻克多模态空间对齐与Sim-to-Real鲁棒性，方能实现真正的无人值守野外作业。

## 英文摘要 (Abstract)

Autonomous navigation in diverse real-world environments remains one of the greatest challenges in robotics. It demands both robust low-level motion control to handle diverse terrain and high-level reasoning to interpret environmental cues. We present ROAM (Robust Open-world Autonomous Mobility), a system that enables legged robots to perform kilometer-scale navigation in complex alpine and urban settings. ROAM combines the world knowledge of a Vision-Language Model (VLM), pre-trained on internet-scale data, with robust locomotion and local planning policies learned in simulation via reinforcement learning (RL). The VLM provides common-sense reasoning and semantic interpretation of the scene, assessing risks, interpreting signs, and producing semantic labels with associated navigation costs. An open-vocabulary segmentation module converts these labels into a dense, real-time risk map, which conditions two pre-trained RL policies responsible for agile locomotion and local navigation, effectively coupling high-level VLM reasoning with low-level physical control. Real-world experiments showed the system's ability to autonomously navigate a wide range of complex real-world environments. The robot completes kilometer-long hikes with minimal intervention, interprets signs to avoid danger, and evaluates its environment to identify safe areas on which to step. This work represents a step toward the reliable deployment of systems in unstructured environments. However, it also underscores the challenges that remain in VLM reliability and spatial reasoning, indicating that foundational models are a crucial component but not yet a complete solution for navigation in the wild.

作者：Takahiro Miki, Robin Schmid, Manthan Patel, Turcan Tuna, Julia Richter, Julian Nubert, Joonho Lee, Matthias M\"uller, Marco Hutter
:::

## 原始参考来源 (Sources)

- [arXiv 2611.25215：Bridging vision-language reasoning and learned control for legged navigation in the wild](https://arxiv.org/abs/2611.25215)