Abstract:This paper presents MuGen, a data-driven framework for learning and deploying multi-skill locomotion on humanoid robots. MuGen enables a robot to perform expressive motions like humans under the guidance of example motion sequences. To achieve this, we employ vector-quantized autoencoders (VQ-VAEs) trained with model-based reinforcement learning, resulting in a generative representation of locomotion that captures key patterns of human motion from hours of heterogeneous human performance data. We employ a teacher-student learning framework and develop a new policy distillation strategy to enable a deployable student policy learning this efficient latent representation. This policy allows the robot to track and mimic unseen human motions and further enables the robot to reuse the learned latent space for other tasks. We demonstrate the effectiveness of our framework through a diverse set of motions and accurate execution.
| Subjects: | Robotics (cs.RO) |
| Cite as: | arXiv:2605.24592 [cs.RO] |
| (or arXiv:2605.24592v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2605.24592 arXiv-issued DOI via DataCite (pending registration) |
From: Yusen Feng [view email]
[v1]
Sat, 23 May 2026 14:06:06 UTC (8,700 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。