Skip to content

论文HF Daily Papers · arXiv 2610.413549 天前

OneSteer单步扩散全身控制实现宇树G1零样本动态避障 ​

本文由 AI 根据公开内容整理,可能有误,请以原文为准:OneSteer: Unlocking Steerable General Humanoid Motion Tracking via One-Step Diffusion Whole-Body Control(HF Daily Papers · arXiv 2610.41354)

核心内容 ​

针对人形机器人部署目标变更需重规划轨迹的痛点,研究团队提出OneSteer单步扩散全身控制器,结合通用运动跟踪与选择性测试时转向能力。该方法在AMASS和BONES数据集上实现最低局部跟踪误差;仿真环境中,单一冻结蒸馏控制器适应多样部署目标,将1.25m低杆清洁通过率从基线的2%提升至90%。

在Unitree G1物理验证中,OneSteer完成9种全身动作序列共27次名义跟踪试验且无摔倒。系统利用机载LiDAR构建符号距离场(SDF)作为转向代价,成功在5次尝试中4次穿过1.1m低杆,3次跨越12cm障碍均未跌倒。此外,基于预测方差的失败评分机制比5次仿真Rollout快10倍以上,可指导定向微调。

关键看点 ​

  • 测试时转向:冻结生成器权重,通过可微任务代价优化输入噪声并结合身体组参考掩码,松弛特定跟踪约束以适应新目标
  • 实机鲁棒性:Unitree G1完成27次名义跟踪零摔倒,仅靠机载LiDAR构建SDF代价,1.1m低杆通过率80%、12cm跨栏成功率100%
  • 高效失败检测:利用模型预测方差对跟踪失败概率排序,无需物理Rollout,评分速度比5次仿真评估快10倍以上并支持定向微调
  • 仿真性能跃升:单一冻结蒸馏控制器适配多目标,1.25m低杆清洁通过率从2%提升至90%,验证了单步扩散架构的泛化能力

工程与产业观察 ​

OneSteer验证了单步扩散模型在具身控制中兼顾通用跟踪与实时避障的可行性。其冻结权重下的噪声优化与SDF代价耦合机制,解决了人形机器人非结构化环境中的零样本适应难题。预测方差替代物理仿真的加速策略,为端侧部署与安全评测提供了低成本工程范式。

英文摘要 (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

Hugging Face Daily Papers 热度:13 upvotes 作者:yifei, Yazhao Yang, Li Weijia, Tianle Shi, Ji Jin, Zhihui Cai, Zihao Liu, Jianzhuang Zhao, Songlin Dong, Xinyuan Chang, Yihong Gong, Dian Shen, Zhiheng Ma

基于本页技术方向关联的整机参数、核心零部件、实测报告与同赛道选型横评。

外部来源

本页事实对应的公开页面,以原文为准。(点击将离开本站)

  1. HF Daily Papers · arXiv 2610.41354:OneSteer: Unlocking Steerable General Humanoid Motion Tracking via One-Step Diffusion Whole-Body Control ↗访问 2026-11-17