CSF: Contextual Safety Filtering for Motion Generators

Lizhi Yang ,

Yiling Hou ,

Yao Tang ,

Junheng Li ,

Daniel Weng ,

Blake Werner ,

Aaron D. Ames

Embodied AI and Robotics DMPA RC AI Safety / AI Ethics AVAR Other AI Safety

基于文本条件的运动生成器能够生成可追踪的全身运动，但其本身并不具备对场景依赖性安全性的认知能力：同一动作可能指向某个物体，也可能指向某个人。现有的安全防护机制，要么仅检查文本提示（prompt），要么依赖带标签的运动数据，要么强制施加几何约束；因此，这些方法均无法直接反映场景上下文如何改变某一运动行为的实际含义。我们提出了“上下文感知安全过滤器”（Contextual Safety Filtering, CSF）——一种无需额外训练的安全过滤机制，它将自然语言描述的安全规则，锚定于该生成器自身所产出的安全与不安全参考轨迹之上。针对每一条正在生效的安全规则，其对应的安全与不安全参考轨迹共同定义了一个仿射形式的安全度量值；该值由一种面向安全参考轨迹跟踪的控制屏障函数-二次规划器（CBF-QP）予以强制保障。我们在四种架构各异、且均已预训练完成的运动生成器上验证了CSF的有效性：结果表明，CSF在所有显式指定的及由场景触发的不安全情形中，均能准确激活相应安全规则，并将危险事件发生率最高降低达90%，同时保留了88%至100%的良性运动。我们进一步在真实世界的Unitree G1人形机器人平台上完整部署并演示了该系统，结果表明，该系统能在包括人机交互、人—物交互在内的多种复杂场景中，成功阻止各类不安全运动的发生。

Text-conditioned motion generators produce trackable whole-body motion, but they have no notion of scene-dependent safety: the same action may target an object or a person. Existing safeguards either inspect the prompt, require labeled motion data, or enforce geometric constraints; therefore, they do not directly account for how scene context changes a motion's meaning. We introduce contextual safety filtering (CSF), a training-free filter that grounds natural-language safety rules in safe and unsafe reference trajectories produced by the generator. For each active rule, safe and unsafe reference trajectories define an affine safety value that a safe reference tracking CBF-QP enforces. Across four pretrained generators with different architectures, CSF activates the intended rules in all explicit and scene-triggered unsafe cases and reduces the danger-event rate by up to 90%, while preserving 88-100% of benign motions. We demonstrate the complete system on a real-world Unitree G1, where it successfully prevents unsafe motions in a variety of scenarios, including interactions with humans and objects.

许愿