Perceptive Humanoid Motion from Simulation
Perceptive Humanoid Motion from Simulation
This research presents HALO, a humanoid autonomy system that overcomes the data bottleneck for generalist robots by synthesizing vision-language-motion datasets across procedurally generated environments. The system learns diverse physical skills such as sitting, navigating to named objects, and pushing boxes in simulation, then transfers these capabilities zero-shot to real-world settings without ever having seen a real home during training. HALO achieves a 91% success rate across 360 trials in six unmapped real-world environments, demonstrating effective sim-to-real transfer for composite, long-horizon tasks.
The paper introduces HALO, a system that synthesizes vision-language-motion data in simulation to enable zero-shot transfer to real-world environments without prior exposure to physical homes during training.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
trials
360 trials
Reported trials
View original evidence
objects, stepping sideways through half-open doors, crouching and crawling beneath tables, and pushing boxes. It orchestrates these skills to solve composite, long-horizon tasks. Without ever seeing a real home during training, HALO transfers zero-shot to the physical world, achieving a 91% success rate across 360 trials in six unmapped real-world environments. See https://apple-aiml-research.github.io/ml-halo/ for c
Open source S4Reported success rate
91%
Reported zero-shot result; the excerpt does not specify an evaluation protocol.
View original evidence
objects, stepping sideways through half-open doors, crouching and crawling beneath tables, and pushing boxes. It orchestrates these skills to solve composite, long-horizon tasks. Without ever seeing a real home during training, HALO transfers zero-shot to the physical world, achieving a 91% success rate across 360 trials in six unmapped real-world environments. See https://apple-aiml-research.github.io/ml-halo/ for c
Open source S4
- paper: not_reported
- code: not_reported
- weights: not_reported
- dataset: not_reported
What remains unknown
Not established in the collected evidence: Environment, Control, Data origin.
Reported performance applies to the described task. It does not establish general autonomy or deployment readiness.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
objects, stepping sideways through half-open doors, crouching and crawling beneath tables, and pushing boxes. It orchestrates these skills to solve composite, long-horizon tasks. Without ever seeing a real home during training, HALO transfers zero-shot to the physical world, achieving a 91% success rate across 360 trials in six unmapped real-world environments. See https://apple-aiml-research.github.io/ml-halo/ for c
Open source S4
ation and wearable capture devices, which require at least an hour of human time to produce an hour of data. We present HALO, a humanoid autonomy system powered by a pipeline that synthesizes vision-language-motion data across a thousand procedurally generated homes. The corpus scales with compute rather than human labor. HALO acquires a repertoire of physical skills, including sitting on chairs, navigating to named
Open source S3
arXiv:2611.06908v1 Announce Type: new Abstract: 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. Some approaches rely on human teleoper
Open source S2
Source：arXiv Robotics — research abstracts · arxiv.org