Abstract:Imitation Learning from monocular video demonstrations provides a scalable approach for teaching complex skills to humanoid robots. However, translating human motion to humanoids requires overcoming significant morphological mismatches. Standard approaches rely on Geometric Retargeting or Indirect Dynamic Retargeting pipelines. We identify that these intermediate kinematic projections introduce a geometric bias, restricting the search space and yielding suboptimal dynamic behaviors. In this paper, we propose Direct Dynamic Retargeting (DDR), a novel single-stage framework that generates high-fidelity, dynamically feasible trajectories directly from expert videos. By formulating the problem in the task space and leveraging a sampling-based Model Predictive Control solver within a physics simulator, DDR natively optimizes over complex contact sequences while mitigating input drift. Our experiments demonstrate that bypassing the geometric bias allows DDR to outperform state-of-the-art baselines in demonstration tracking accuracy. Furthermore, we establish that providing such physically viable references to RL agents accelerates training convergence and enhances the final execution of agile and balancing behaviors. Source code will be made publicly available.
| Subjects: | Robotics (cs.RO) |
| Cite as: | arXiv:2605.23762 [cs.RO] |
| (or arXiv:2605.23762v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2605.23762 arXiv-issued DOI via DataCite (pending registration) |
From: Constant Roux [view email]
[v1]
Fri, 22 May 2026 15:33:40 UTC (19,412 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。