Safety layer for VLA models and humanoid robots. Sub-ms physics guard, ISO-aligned, zero neural nets.
-
Updated
Aug 12, 2026 - Python
Safety layer for VLA models and humanoid robots. Sub-ms physics guard, ISO-aligned, zero neural nets.
Deterministic physics safety for embodied AI — beyond statistical safety guards.
Physics-based collision detection with dynamic contact area, impulse boundary, and 7-level risk grading.
Deterministic Newtonian safety engine for robotics — force, speed, contact area, all computed.
500+ labeled VLA safety scenarios across 7 risk levels — benchmark dataset for safety evaluation.
Add a description, image, and links to the impulse-safety topic page so that developers can more easily learn about it.
To associate your repository with the impulse-safety topic, visit your repo's landing page and select "manage topics."