外观
论文arXiv 2610.1899113 天前
Contact-UAN残差世界模型修正执行器动力学实现人形机器人力控迁移
本文由 AI 根据公开内容整理，可能有误，请以原文为准：Contact-UAN: Unsupervised Actuator Networks as a Residual World Model for Humanoid Control（arXiv 2610.18991）
核心内容
人形机器人仿真策略迁移硬件时常因执行器Sim-to-Real差距导致力交互失效，影响扭矩效率与抗冲击鲁棒性。研究团队提出Contact-UAN残差世界模型，在保留仿真物理先验的同时修正执行器动力学，使仿真回放与硬件记录对齐。该方法仅需6.5分钟真实行走数据训练，即可泛化至需精确力/冲量调节的任务，实现策略无缝迁移。
技术核心在于接触一致性回放机制：当硬件回放初始接触 wrench 未对齐时，模型易学习重置伪影而非执行器差异。Contact-UAN仅依赖本体感知推断仿真侧立足足姿态，确保回放初始化力一致。实测表明，该模型使真机运行扭矩降低20%、力跟踪误差减少43%，并具备极致柔顺性、准确阻抗控制能力及高鲁棒性动态跳跃表现，有效解决了纯仿真训练在力控层面的迁移瓶颈。
关键看点
- 残差建模：保留仿真物理先验并修正执行器动力学，解决力交互迁移失效与扭矩效率低问题
- 数据高效：仅利用6.5分钟真实世界行走数据训练UAN，即可泛化至复杂力控与动态跳跃任务
- 性能提升：真机部署后运行扭矩降低20%，力跟踪误差减少43%，实现精准阻抗控制
- 接触对齐：仅凭本体感知推断仿真侧立足姿态，消除回放初始wrench错位导致的模型学习偏差
工程与产业观察
该工作验证了基于少量本体感知数据的残差世界模型是弥合人形机器人Sim-to-Real力控鸿沟的有效路径。通过解耦物理先验与执行器误差，大幅降低了高动态任务对大规模真机力控数据的依赖，为具身智能从运动模仿迈向精密力操作提供了低成本工程范式。
英文摘要 (Abstract)
Humanoid policies trained in simulation can transfer qualitatively to hardware while failing to transfer the force interactions learned during training, compromising torque efficiency, load regulation, and robustness to impacts. This failure can arise from actuator sim-to-real gaps because forces depend on actuator torques. We introduce Contact-UAN, a residual world model that preserves a simulator's physics priors while correcting actuator dynamics to align simulated rollouts with recorded hardware rollouts. For humanoid motion with contact, this requires contact-consistent replay: if hardware playback in simulation begins with misaligned contact wrenches, the residual model learns to correct reset artifacts rather than actuator discrepancies. Using only onboard proprioception, Contact-UAN infers the simulator-side stance-foot pose needed to initialize replay with consistent contact forces. We show that a UAN trained on 6.5 minutes of real-world walking data generalizes to tasks that benefit from accurate force and impulse regulation and produces policies that transfer seamlessly to the real world, yielding 20% lower running torque, 43% lower force-tracking error, extreme compliance, accurate impedance control, and dynamic, robust jumps.
作者：Nolan Fey, Khai Nguyen, Yash Prabhu, Gabriel B. Margolis, Pulkit Agrawal
来源
本页事实对应的公开页面，以原文为准。