arXiv：cs.LG· Derek You, Zafir Shamsi, Keqin Wang, Christine Allen-Blanchette·· 4 小时前
高阶形态先验如何提升执行器退化下的四足机器人强化学习
Higher-Order Morphology Priors for Quadruped Reinforcement Learning Under Actuator Degradation
AI 导读
研究将 Unitree Go1 表示为带肢体与身体级 rank-2 单元的胞腔复形，并采用 Hodge 消息传递建模高阶机械结构。在执行器退化训练下，node-edge-face Hodge actor 在未见过的执行器退化情形中取得最高回报，存活率更高且速度跟踪误差更低，表明高阶形态可作为全身补偿的有效归纳偏置。
正文
Abstract:Actuator degradation turns quadruped locomotion into a coordination problem requiring joints to compensate for lost actuation. Prior work suggests that morphology-aware graph policies improve learning and generalization under body perturbations. We ask whether these benefits can be strengthened by explicitly modeling higher-order mechanical structure. We represent the Unitree Go1 as a cell complex with limb- and body-level rank-2 cells and apply Hodge-based message passing. Under degradation training, the node-edge-face Hodge actor achieves the highest return on unseen actuator degradations, with higher survival and lower velocity-tracking error. These results support higher-order morphology as a useful inductive bias for whole-body compensation under actuator degradation.
| Comments: | Accepted to IROS Workshop BLPC 2026 |
| Subjects: | Robotics (cs.RO); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.10934 [cs.RO] |
| (or arXiv:2610.10934v1 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2610.10934 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Keqin Wang [view email]
[v1]
Wed, 7 Oct 2026 21:42:18 UTC (2,244 KB)
来源：arXiv：cs.LG · arxiv.org