Training a 128k-param neural cellular automata to 'see around corners' with sound
yacineMTB · x · 2026-10-04
Developer movaxbx is training a tiny 128k-parameter neural cellular automata (NCA) to produce a location probability map in a simulated acoustic environment — teaching an AI to 'see around corners' with sound.
In the visualization, the purple circle is the model's prediction, the green dot the target, and blue the simulated binaural receiver. The author notes it's a fun return to NCAs after several months — a small-scale experiment in localizing sound sources from binaural audio.
More from Embodied
- CoreWeave's Physical AI Lead Explains Why Robots Hit the Sim-to-Real Gap — thursdai_pod · 2026-10-04
- Replacing CAD with code: LLM-written build123d Python scripts design entire robots — natesiggard · 2026-10-03
- HUST, PKU and Keenon Robotics unveil AdaRoboVLG for context-aware robot grasping — jiqizhixin · 2026-10-03
- Sharpa D01 humanoid robot outperforms humans at grasping soft dropped objects — CyberRobooo · 2026-10-03
- The $899 Mac Mini M6 was built for 24/7 agentic computing — most owners use it to check email — hey_abusiddik · 2026-10-03
- Figure AI retires F.02 humanoid fleet by training robots to leap into molten steel — heatherknight · 2026-10-03