外观
论文arXiv 2610.1563917 天前
通过离散与节律 Primitive 组合进行模仿学习
原文链接：Imitation Learning via Discrete and Rhythmic Primitive Composition（arXiv 2610.15639 · 发布日期：2026-10-11）
Imitation Learning via Discrete and Rhythmic Primitive Composition
针对长时程机器人操作的子任务组合问题，提出NDC用原始轨迹监督组合DMPs，并在2D handwriting和3D robotic tasks上验证。
核心要点
- 长时程机器人操作由语义子任务组成，现有MP组合依赖人工设计或样本低效强化学习。
- NDC在仅给定每个primitive单个示例时，用原始轨迹监督组合DMPs。
- 方法以DMP并行组合松弛类别选择，并用warp-invariant loss处理时空变化。
- 在2D handwriting和3D robotic tasks上，多数任务和指标优于基线，消融显示各设计贡献。
验证范围与局限：验证范围为2D handwriting和3D robotic tasks，摘要未说明是否为真机实验。
英文摘要 (Abstract)
Long-horizon robotic manipulation is inherently composed of semantically meaningful sub-tasks. While Movement Primitives (MPs) can model these components, their composition typically relies on manual engineering or sample-inefficient reinforcement learning. Existing supervised methods are limited by the lack of MP labels in standard datasets and often fail to generalize across discrete and rhythmic movements. We propose Neural Dynamic Composition (NDC), a supervised framework to compose Dynamic Movement Primitives (DMPs) using only raw trajectories for supervision, given a single example of each primitive. Our approach exploits parallel combinations of DMPs as a differentiable relaxation of categorical primitive selection, and utilizes a warp-invariant loss formulation to identify primitives despite spatial or temporal variations. A staged training curriculum is employed to learn the selection policy and DMP parameters, unifying discrete and rhythmic movements in one framework. Evaluated on 2D handwriting and 3D robotic tasks, our method outperforms baselines on most tasks and metrics. An ablation study demonstrates how each design choice contributes to the system's performance.
作者：Rafid Saif, Auke Jan Ijspeert, Hsiu-Chin Lin
来源
本页事实对应的公开页面，以原文为准。