# Workhorse robot learns whole-body tasks from human demos

> Researchers trained a Unitree G1 to sort boxes and recover from pushes using human‑recorded motions.

Oossa · 2026-10-08 · https://oossa.com/en/workhorse-robot-learns-whole-body-tasks-from-human-demos

The team behind Workhorse taught a Unitree G1 humanoid to plan and execute whole‑body manipulation by watching egocentric video of people. A visual planner predicts poses for the torso, wrists and feet, while a reinforcement‑learning tracker follows those targets. On a real robot the system sorts boxes, catches a thrown box and even climbs a suitcase, and it can recover when a person pushes it. In simulation it completed box‑sorting in 77% of runs and 64% when subjected to pushes up to 40 N·s.

## The facts

- 77% success rate in simulated box‑sorting episodes
- 64% success under pushes of up to 40 N·s

## Why it matters

If the approach scales, everyday service robots could handle messy, contact‑rich tasks without hand‑crafted programming.

## Sources & references

1. [Workhorse: Learning Robust Whole-Body Humanoid Loco-Manipulation from Human Data](https://arxiv.org/abs/2610.09117) – arXiv cs.RO, 2026-10-08

Last updated: 2026-10-08