论文arXiv 2611.149859 天前
ILA适配器实现人形机器人无传感器柔性运动跟踪与策略迁移
本文由 AI 根据公开内容整理,可能有误,请以原文为准:ILA: A Transferable Interface-Level Adapter for Compliant Humanoid Motion Tracking(arXiv 2611.14985)
核心内容
针对人形机器人在受限环境或人机交互中直接跟踪参考动作存在的安全风险,研究提出接口级适配器ILA(Interface-Level Adapter)。该方案旨在解决传统柔性跟踪依赖昂贵力/力矩传感器的问题,通过在预训练运动跟踪策略之上增加轻量级适配层,使机器人能够响应外部物理引导。实验表明,ILA在三款不同的人形运动跟踪器上均实现了63.6%至75.4%的引导目标误差降低,且硬件验证证实了其在保持原有跟踪性能的同时显著提升了对人类引导的响应速度。
技术上,ILA作为独立模块从机器人观测历史中自适应调整运动参考指令,预测由物理交互引起的参考修正量。其核心优势在于“接口级”设计:适配器完全在仿真环境中训练,部署时可直接插入不同的预训练跟踪策略中,无需对适配器本体或底层跟踪策略进行任何重训练。这种解耦架构使得柔性引导能力成为可复用的通用扩展组件,大幅降低了为现有人形机器人系统增加安全交互能力的工程门槛与算力成本。
关键看点
- 无感柔性交互:仅利用本体观测历史推断物理引导意图,无需集成专用力/力矩传感器,显著降低硬件BOM成本与系统集成复杂度。
- 零样本策略迁移:采用接口级解耦设计,适配器在仿真训练后可直接即插即用至不同预训练跟踪器,无需微调底层策略或重训适配器。
- 量化性能提升:在三款异构人形运动跟踪策略上实测,引导目标误差较原始策略降低63.6%~75.4%,同时保持标称运动跟踪精度不衰减。
工程与产业观察
ILA验证了通过轻量级适配层将“柔性交互”从硬件依赖转化为算法插件的可行性。对于具身智能开发者,这意味着可在不改动底层控制栈、不增加力传感器的情况下,快速为现有预训练人形机器人补齐安全示教与物理协作能力,极大降低了Sim-to-Real阶段的人机交互调试成本与数据迭代周期。
英文摘要 (Abstract)
Humanoid motion tracking is fundamental to reproducing human motions on humanoid robots and serves as an important tool for teleoperation and motion data collection. However, directly tracking reference motions can be unsafe in constrained environments or during physical human--robot interaction, motivating the need for compliant motion tracking. We present ILA, an Interface-Level Adapter for compliant humanoid motion tracking without dedicated interaction sensors. ILA adapts motion references from the robot's observation history, allowing the robot to respond to physical guidance while preserving the capabilities of a pretrained motion tracker. As an interface-level module, ILA can be directly plugged into different pretrained motion trackers without retraining either the adapter or the tracking policies. ILA is trained entirely in simulation to predict reference corrections induced by physical interaction. We evaluate ILA with three humanoid motion trackers and achieve a 63.6--75.4% reduction in guidance-target error compared with their original policies. Matched simulation studies and hardware experiments further show that ILA improves responsiveness to human guidance while preserving nominal motion-tracking performance. These results demonstrate compliant guidance as a reusable, sensor-free extension to pretrained humanoid motion trackers.
作者:ZhuoChang Cai, Zihao Yan, Jinxuan Zhu, Yanjie Li, Hao Xiong
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)