论文/arXiv 2610.12440·20 天前

面向人到机器人灵巧操作的生成式神经重定向

针对人类示教因具身差异难以直接用于机器人，GNR用流匹配采样可行轨迹，并在灵巧操作重定向与real-to-sim数据生成中验证。

AI 整理

• 本体 SDK 与控制
• 数据与采集
• 具身模型

原文链接：Generative Neural Retargeting for Human-to-Robot Dexterous Manipulation（arXiv 2610.12440 · 发布日期：2026-10-08）

Generative Neural Retargeting for Human-to-Robot Dexterous Manipulation

针对人类示教因具身差异难以直接用于机器人，GNR用流匹配采样可行轨迹，并在灵巧操作重定向与real-to-sim数据生成中验证。

核心要点

• 人类示教易扩展，但具身差异使动作不能直接在机器人上执行。
• GNR假设可行轨迹位于共享低维流形，用流匹配模型条件采样。
• 相比MPC，GNR仅用其8.5%样本，成功率为56.20%对27.20%。
• 在real-to-sim数据引擎中生成含密集接触力标签的大规模数据集。

验证范围与局限：验证涉及灵巧操作重定向对比和real-to-sim数据生成，摘要未说明真机实验。

英文摘要 (Abstract)

Human demonstrations are a scalable data source for learning dexterous manipulation, but the embodiment gap prevents human motion from being executed directly on robots. Inverse kinematics (IK) retargets human motion to robots efficiently but ignores dynamics, often producing infeasible motions. Reinforcement learning (RL) and sampling-based model predictive control (MPC) are commonly employed to yield dynamically feasible motions, but both are sample-inefficient and sensitive to hyperparameters. RL suffers from costly and unstable training and tedious reward engineering; MPC avoids policy optimization, yet retargets each trajectory in isolation, and solving one does not make the next easier. Sampling cost grows rapidly with dataset size and task difficulty. We hypothesize that dynamically feasible trajectories concentrate near a low-dimensional manifold shared across demonstrations, so that retargeting can be reduced to sampling from that manifold, conditioned on human motion, rather than solving a fresh optimization problem for every demonstration. We propose \textbf{Generative Neural Retargeting} (GNR), which uses a flow matching model to sample feasible trajectories. GNR outperforms MPC with only $8.5%$ of the samples required by MPC, achieving a success rate of $56.20%$ compared to $27.20%$ for MPC. GNR can be used for scalable and efficient retargeting of large-scale, long-horizon, and millimeter precision human demonstrations: by applying GNR within a real-to-sim data engine, we produce a dexterous manipulation dataset with dense contact-force labels, spanning $223$k demonstrations and $3.3$k object geometries.

作者：Dechen Gao, Yang Yue, Ben Abbatematteo, Nathan Godwin, Pengcheng Wang, Roger Boldú, Steven Man, Zhiyang Dou, Chuan Qin, Sho Nakagome

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.12440：Generative Neural Retargeting for Human-to-Robot Dexterous Manipulation访问 2026-10-21