# ResGAC improves end‑effector tracking on Unitree G1 humanoid

> The new ResGAC controller cuts pose errors and raises peg‑in‑hole success from 50% to 90% on a standing humanoid.

Oossa · 2026-10-08 · https://oossa.com/en/resgac-improves-end-effector-tracking-on-unitree-g1-humanoid

Researchers led by Joohwan Seo introduced ResGAC, a whole‑body controller for humanoid robots. It blends geometric admittance control with a residual reinforcement‑learning layer to cancel unmodelled dynamics. Tested on a real Unitree G1, ResGAC reduced translation and rotation errors compared with the prior SONIC baseline. In a standing peg‑in‑hole trial it succeeded 90% of the time, versus 50% for SONIC.

## The facts

- Paper published on 2026-10-08
- ResGAC achieved 90% success in standing peg‑in‑hole, SONIC 50%

## Why it matters

More accurate tracking lets humanoid robots perform delicate tasks while walking, expanding their use in manufacturing and service settings.

## Sources & references

1. [Precise SE(3) End-Effector Tracking in Whole-Body Humanoid Control](https://arxiv.org/abs/2610.09479) – arXiv cs.RO, 2026-10-08

Last updated: 2026-10-08