Skip to content

论文arXiv 2609.205661 个月前

OmniMimic框架实现四足机器人全向多步态运动增强与策略迁移 ​

本文由 AI 根据公开内容整理,可能有误,请以原文为准:OmniMimic: Dynamics-completed Motion Augmentation for Multi-style Omnidirectional Quadruped Locomotion(arXiv 2609.20566)

核心内容 ​

针对动物运动演示在后退、侧移及转向等非前向维度缺乏风格监督的难题,研究提出OmniMimic训练框架。该方法将方向受限的动物演示转化为覆盖目标轴速度范围的统一多步态策略,旨在解决四足机器人自然步态迁移中的方向性缺失问题。在四种步态的仿真测试中,OmniMimic相比匹配的APEX基线,将均匀笛卡尔指令网格上的速度跟踪RMSE降低了63.1%,前后向参考速度下的平均足端位置RMSE降低了12.9%。

技术上,OmniMimic结合时间反转、约束动力学补全和矢状面反射,构建超出观测方向的机器人专用运动学与物理监督信号。系统采用渐进式指令扩展策略,从演示速度分布逐步逼近目标边界;控制架构使用共享Actor配合软门控步态专用残差专家(Residual Experts),在复用通用运动技能的同时进行步态特异性修正。这种机制确保了机器人在全向速度空间内既能保持动物演示的自然风格,又能精确响应复杂运动指令。

关键看点 ​

  • 数据增强:融合时间反转、约束动力学补全及矢状面反射,生成超越原始观测方向的运动学与物理监督信号
  • 网络架构:采用共享Actor加软门控步态专用残差专家结构,平衡通用运动技能复用与特定步态的风格修正
  • 性能指标:仿真中速度跟踪RMSE较APEX基线降低63.1%,前后向足端位置RMSE降低12.9%
  • 指令扩展:实施渐进式命令扩展策略,从有限演示速度分布平滑过渡至目标全轴向速度边界

工程与产业观察 ​

该工作为具身智能运动模仿提供了关键的数据工程范式,证明了通过算法层面的动力学补全与结构化网络设计,可突破生物演示数据的物理采集限制。对于四足机器人研发,这意味着无需大规模重新采集全向数据即可复用高质量前向演示,显著降低了自然步态策略的训练成本与Sim-to-Real迁移门槛。

英文摘要 (Abstract) ​

Animal demonstrations provide quadruped robots with natural and distinctive gait styles that are difficult to specify through hand-crafted rewards. However, their narrow directional coverage leaves little style-consistent supervision for backward, lateral, and turning commands. We present OmniMimic, a training framework that turns directionally limited animal demonstrations into a single multi-gait policy over target per-axis velocity ranges. OmniMimic first combines temporal reversal, constrained dynamics completion, and sagittal reflection to construct robot-specific kinematic and physical supervision beyond the observed directions. It then expands commands progressively from the demonstrated velocity distribution toward the target per-axis bounds, and uses a shared actor with soft-gated, gait-specialized residual experts to balance reusable locomotion skills with gait-specific corrections. Across four gaits in simulation, OmniMimic reduces mean foot-position RMSE at forward and backward reference velocities by 12.9% and velocity-tracking RMSE on a uniform Cartesian command grid by 63.1%, compared with the matched APEX baseline. The project page is at https://OmniMimic.github.io.

作者:Sheng Wu, Guoqiang Zhao, Zhe Yang, Fei Teng, Zhikun Zhou, Yanlin Yang, Zheng Fang, Hong Zheng, Yaonan Wang, Kailun Yang

来源

本页事实对应的公开页面,以原文为准。

  1. arXiv 2609.20566:OmniMimic: Dynamics-completed Motion Augmentation for Multi-style Omnidirectional Quadruped Locomotion访问 2026-10-12