Ken Goldberg's Agentic Robotics: AI writes deterministic robot code, no demos needed
animesh_garg · x · 2026-09-19
Berkeley professor Ken Goldberg and collaborators argue a paradigm shift is underway in robotics: the bottleneck isn't 100,000 hours of human demonstrations but software engineering.
Their "Agentic Robotics" path uses zero human demonstration data, multi-agent coding harnesses (GaP), video-to-sim system identification, and offline agent self-correction deployed as lightweight deterministic edge code — AI builds the robot's code rather than controlling its inner loop.
More from Embodied
- Baidu's Apollo Go wraps two-week public robotaxi test rides in Tung Chung, Hong Kong — Baidu_Inc · 2026-11-04
- imec neural probe integrates photonics directly into 130nm CMOS backend — jwt0625 · 2026-11-04
- Chris Paxton: interactive robot sim is the future but glaringly data-limited — Scobleizer · 2026-11-04
- DreamX: video diffusion models as universal latent action learners for robots — kaist-ai · 2026-11-04
- RoboTryOn turns human videos into robot demos, lifting success rate from 2% to 60% — kaist-ai · 2026-11-04
- INMO Future glasses: dual 1080p displays, 125g, $799 early-bird on Kickstarter Dec 10 — Scobleizer · 2026-11-04