论文/arXiv 2610.02341·27 天前

面向安全人形机器人全身跟踪的滤波器感知微调

针对人形机器人安全全身跟踪，提出滤波器感知微调CoFiT，并在多约束场景与Unitree G1真机上验证。

AI 整理

• 硬件零部件
• 本体 SDK 与控制
• 数据与采集

宇树 G1

原文链接：Filter-Aware Fine-Tuning for Safe Humanoid Whole-Body Tracking（arXiv 2610.02341 · 发布日期：2026-10-01）

Filter-Aware Fine-Tuning for Safe Humanoid Whole-Body Tracking

针对人形机器人安全全身跟踪，提出滤波器感知微调CoFiT，并在多约束场景与Unitree G1真机上验证。

核心要点

• 独立处理跟踪策略与安全滤波器会因动作和状态分布改变而产生不匹配。
• 论文通过案例研究分离动力学、目标和信息不匹配，并据此提出CoFiT。
• 多种约束场景中，CoFiT较仅用滤波训练在TWIST2/SONIC将违规时间降91%/21%。
• Unitree G1上TWIST2违规时间降83%，且所有试验无需操作员干预。

验证范围与局限：验证范围为多种约束场景、TWIST2/SONIC评估及Unitree G1硬件全身跟踪试验。

英文摘要 (Abstract)

Safe whole-body motion is essential for deploying humanoid robots in unstructured environments. Modern humanoid control commonly separates reference specification from execution, with a planner, teleoperator, or motion generator providing a reference that a reinforcement-learning policy tracks through dynamically feasible whole-body control. Runtime safety filters, such as control barrier functions (CBFs), offer a promising approach for enforcing newly introduced constraints via interventions on the tracker's outputs. We show, however, that treating the tracking policy and safety filter independently induces fundamental mismatches, as filtering alters both the executed actions and the induced state distribution. We study this policy-filter interface through case studies that isolate dynamics, objective, and information mismatches, highlight their root causes, and use these insights to develop CoFiT (Constrained Filter-aware Tuning), a filter-aware fine-tuning method for pretrained trackers. Across diverse constraint scenes, CoFiT reduces violation time relative to filter-only training by 91% on TWIST2 and 21% on SONIC, while requiring smaller safety filter corrections. On Unitree G1 hardware, CoFiT reduces violation time by 83% for TWIST2 and completes every trial without operator intervention, whereas 50% of baseline trials require an operator stop. Together, these results provide actionable insights into policy-filter interactions and establish design principles for integrating learned trackers with runtime safety filters.

作者：Pranit Mohnot, Christian Helten, Daniele Gammelli, Marco Pavone

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.02341：Filter-Aware Fine-Tuning for Safe Humanoid Whole-Body Tracking访问 2026-10-21