Blog explains how SO-100 robot arms are controlled, written for ML practitioners
KyleMorgenstein · x · 2026-09-20
- Jacob Rothschild published a blog post, How robot arms are controlled, aimed at ML practitioners entering robot learning, explaining how an SO-100 robot arm is controlled to help them get past the hurdle of directly controlling hardware.
- It is the first in a series of explainers from his own speedrun, with a follow-up on ACT (Action Chunking with Transformers) promised next.
- Control researcher JieWangZJUI, who reshared it, said he's "tired of teaching ML folks control theory" and sees the blog as filling that knowledge gap.
Related event: ML Practitioners Get a Hands-On Intro to Robot Arm Control(3 posts)→
More from Embodied
- Musk confirms every Starlink V3 satellite will carry an Nvidia Vera Rubin NVL72 — 100k sats, 25GW — ns123abc · 2026-09-20
- Odyssey-3: One Pretrained World Model Adapts to Different Robot Arms With Hours of Demos — ChongZzZhang · 2026-09-20
- iPhone 18 Pro motherboard weighs just 13.1g yet packs PC-class compute — SumitGup · 2026-09-20
- Snap teams with Salesforce, Amazon and Nvidia to push Specs glasses into the enterprise — matt_slotnick · 2026-09-20
- Innate bot demo: founder shows robot running, dubbed 'Apple II of robots' — Scobleizer · 2026-09-20
- VLA-Replica: a $-low-cost real-world VLA benchmark where NVIDIA GR00T N1.7 matches π₀ with 50 demos — YuXiang_IRVL · 2026-09-20