Neuro-Symbolic VLM for Adaptive Impedance Control
Connecting a VLM's scene understanding to a low-level impedance controller
A system that connects a VLM’s scene understanding to a low-level impedance controller for a Franka Emika arm, validated and tested in NVIDIA Isaac Sim.
- Material classification accuracy from 28% (pure visual) to 100%, by combining OpenCV geometric detection with VLM reasoning over scene graphs.
- Aimed at safer human-robot interaction.