ETH's AME-2 Legged Locomotion Paper Accepted by TRO, G1 Training Code Open-Sourced

ETH Zurich's AME-2 paper on attention-based neural map encoding for agile and generalizable legged locomotion has been officially accepted by IEEE TRO. The accompanying training code for Unitree G1 has been open-sourced.

2026-10-06 ~ 2026-10-06 · 2 related posts

• ETH's AME-2 legged locomotion paper accepted at TRO, training code open-sourced — ChongZzZhang · 2026-10-06
• G1 training codebase released as robot learning paper accepted at TRO — ChongZzZhang · 2026-10-06