OneSteer: Unlocking Steerable General Humanoid Motion Tracking via One-Step Diffusion Whole-Body Control
Abstract
Deployment objectives often specify desired outcomes without providing revised trajectories, while parts of the original reference must still be followed. We present OneSteer, a one-step diffusion whole-body controller that combines general motion tracking with selective test-time steering. To accommodate new objectives, differentiable task costs guide optimization of the generator's input noise with frozen weights, while body-group reference masks relax selected tracking requirements. Predicted state-action plans provide a shared interface for task optimization and feedback-conditioned execution. For nominal tracking, the model trained from scratch achieves the lowest local tracking error among the evaluated baselines on successful held-out AMASS and BONES clips. For steering, one frozen distilled controller adapts across diverse deployment objectives in simulation, raising the clean passage rate under a 1.25 m bar from 2% to 90%. On a physical Unitree G1, OneSteer completes all 27 nominal tracking trials across nine diverse whole-body motion sequences without falling. With a steering cost derived from a signed distance field built using only onboard LiDAR, it also passes under a 1.1 m low bar in four of five attempts and steps over a 12 cm hurdle without falling in all three attempts. Beyond control, prediction variation in the model trained from scratch ranks likely tracking failures without physics rollouts. This scoring method is more than 10x faster than simulation-based scoring with five rollouts per clip and guides targeted fine-tuning. Project page: https://onesteer.github.io
Community
Thrilled to release OneSteer—enabling training-free, zero-shot real-world adaptation for humanoid motion tracking! 🚀
Standard trackers struggle with unseen obstacles (low bars, hurdles), tray carrying, or sudden motor failures without costly retraining or unnatural motion edits. OneSteer uses a one-step diffusion model to map noise directly into a 1s differentiable "state–action plan." At test time, we simply optimize the input noise via gradients with body-group masking—zero weight updates needed.
📊 Key Highlights:
- SOTA Precision & Agility: 26.5 / 22.6 mm error on AMASS / BONES; 0 falls across 27 Unitree G1 dynamic trials (including backflips).
- Zero-Shot Adaptation: Low-bar (1.25 m) pass rate jumps from 2% → 90%, plus spill-free tray carrying, high-fives, and knee-failure recovery.
- Real-World LiDAR Deployment: Same frozen weights duck under a 1.1 m bar (4/5) and clear a 12 cm hurdle (3/3, 0 falls) while dancing.
- 10× Faster Failure Screening: Multi-modal plan divergence flags risky motions without physics sim (91% precision in top 2%), fixing 48% of base model failures.
🌐 Project Page: https://onesteer.github.io
📄 Paper: https://arxiv.org/abs/2610.41354
Check out the project page for real-world backflip and obstacle-avoidance demos. Feedback and discussions are welcome—thanks for your support! 🙌
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- RedirectMimic: Local Motion Content Redaction for Humanoid Whole-Body Control (2026)
- GRAFT: Geometry-Aware Residual Adaptation for Foundation Trackers (2026)
- METATRON: A Promptable Whole-Body Execution Layer for Humanoids (2026)
- PredActor: Predictive Action Diffusion for Steerable Onboard Humanoid Control (2026)
- SuperMimic: Scaling Humanoid Whole-Body Control with One Differentiable World Model (2026)
- Test-Time Motion Steering for Perceptive Humanoid Fall Recovery (2026)
- AMBIT: Anticipatory Multimodal Body Recruitment for Bimanual Tracking on a Humanoid (2026)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend
Get this paper in your agent:
hf papers read 2610.41354 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper