# 足式机器人VLN安全架构：dEBF约束与SDF感知保障VLA导航避障

> 原文链接：https://robotics.hellobike.com/papers/2026-11-04-2611-04972-b708196b
> 频道：papers | 更新日期：2026-11-04 | 标签：auto, arxiv
> 核心摘要：针对足式机器人VLA模型缺乏物理接地与安全保证痛点，提出Caging VLN架构。通过集成SDF感知、Bezier-MPPI规划器及dEBF约束，在人形与四足机器人实机验证中实现可证明碰撞规避，兼容主流VLM/VLA模型。

# 足式机器人VLN安全架构：dEBF约束与SDF感知保障VLA导航避障

::: v-pre
> 本文由 AI 根据公开内容整理，可能有误，请以原文为准：[Caging Vision-Language Navigation for Legged Robots via Barrier-Certified Safety Architecture](https://arxiv.org/abs/2611.04972)（arXiv 2611.04972）

## 核心内容

针对现有视觉语言动作（VLA）模型在足式机器人导航中缺乏物理接地与显式安全保障的问题，研究提出Caging VLN安全架构。该方案不摒弃端到端VLA规划能力，而是通过增加物理接地层对其进行增强与约束。系统已在人形与四足机器人上完成广泛实机验证，证明了其在复杂场景下兼具通用性、高效性与安全性，且模块化设计使其兼容各类VLM/VLA模型及机器人平台。

技术层面，高层VLA规划器与低层物理接地层协同工作，后者利用高效状态估计、原位3D空间感知及符号距离场（SDF）表征生成可执行轨迹。为确保导航安全，系统采用低维Bezier参数化MPPI规划器，并结合离散指数障碍函数（dEBF）约束，在预设假设下提供可证明的碰撞规避保证。该架构将语义理解与底层安全控制解耦，解决了大模型落地具身智能时的安全边界问题。

## 关键看点

- **安全控制机制**：结合低维Bezier参数化MPPI规划器与离散指数障碍函数dEBF约束，提供理论可证明的碰撞规避保证
- **物理感知表征**：低层物理接地模块集成高效状态估计与原位3D感知，利用符号距离场SDF生成可执行安全轨迹
- **系统兼容架构**：采用模块化设计对端到端VLA进行增强而非替代，支持多种VLM/VLA模型接入且平台无关
- **实机验证范围**：在人形与四足两类足式机器人平台上完成广泛实测，验证了视觉语言驱动导航的安全性与效率

## 工程与产业观察

该研究为具身大模型落地提供了关键安全范式：通过dEBF与SDF构建确定性安全笼，解决VLA模型概率性输出的物理风险。这种“语义规划+安全兜底”的分层架构，平衡了泛化能力与工程可靠性，是足式机器人从仿真走向真实非结构化环境的必经技术路径。

## 英文摘要 (Abstract)

By harnessing the inherent commonsense capabilities of pretrained Large Language Models (LLMs) and Vision-Language Models (VLMs), current Vision-Language Action(VLA) models for legged robots excel at interpreting language-based navigation commands but lack the physical grounding and explicit safety assurance required for real-world operation. Therefore, this paper aims to cage Legged-Vision-Language Navigation (VLN) through a safety architecture that provides physical grounding and ensures navigation safety. The key idea is not to abandon the powerful capabilities of end-to-end VLA planning, but rather to augment and cage them. In particular, we integrate a high-level VLA planner with a low-level physical-grounding layer that leverages efficient state estimation, in-situ 3D spatial perception, and an Signed Distance Field (SDF) representation to generate executable trajectories. We provide a navigation planner with provable collision-avoidance guarantees under the stated assumptions by combining a low-dimensional Bezier-parameterized MPPI planner with Discrete Exponential Barrier Function (dEBF) constraints. The proposed Legged-VLN system is modular, making it compatible with various VLM/VLA models and agnostic to the robot platform. The proposed system is validated with extensive real-world experiments on humanoid and quadrupedal robots, demonstrating versatile, efficient, and safe vision-language driven navigation capabilities.

作者：Abhijeet M. Kulkarni, Saimouli Katragadda, Yuze Du, Zijun Sha, Guoquan Huang
:::

## 原始参考来源 (Sources)

- [arXiv 2611.04972：Caging Vision-Language Navigation for Legged Robots via Barrier-Certified Safety Architecture](https://arxiv.org/abs/2611.04972)