AI TrackerFrontier desk · AEST
Back to news
normalaitech

Perceptive Humanoid Motion from Simulation

arXiv cs.AI · 5 Nov 2026, 04:55 AEST

Humanoid robots are advancing towards generalist autonomy, but face challenges due to a lack of large-scale data. The HALO system synthesizes vision-language-motion data to enable humanoid robots to perform complex tasks in real-world environments.

Key facts

Read aloud

Humanoid robots are advancing towards generalist autonomy, but face challenges due to a lack of large-scale data. The HALO system synthesizes vision-language-motion data to enable humanoid robots to perform complex tasks in real-world environments. Humanoid robots have the potential to automate a wide range of physical tasks. Progress toward generalist humanoid autonomy, however, remains constrained by a data bottleneck: mapping pixels and free-form language to low-level joint commands requires paired supervision, and no corpus of such data exists at sufficiently large scale. We present HALO, a humanoid autonomy system powered by a pipeline that synthesizes vision-language-motion data across a thousand procedurally generated homes. HALO acquires a repertoire of physical skills, including sitting on chairs, navigating to named objects, stepping sideways through half-open doors, crouching and crawling beneath tables, and pushing boxes. Without ever seeing a real home during training, HALO transfers zero-shot to the physical world, achieving a ninety one percent success rate across three hundred sixty trials in six unmapped real-world environments.

Sources in this cluster