期刊
IEEE ROBOTICS AND AUTOMATION LETTERS
卷 7, 期 2, 页码 4488-4495出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/LRA.2022.3150024
关键词
Task analysis; Robots; Adaptation models; Friction; Trajectory; Gears; Couplings; Learning from demonstration; reinforcement learning; sensorimotor learning
类别
资金
- EPSRC iCASE Award
- Thales Maritime Systems
- EPSRC U.K. RAI Hub NCNR [EPR02572X/1]
- Alan Turing Institute, as part of the Safe AI for surgical assistance project at the University of Edinburgh
- Google X AI Residency
- Academy of Finland Flagship Programme: Finnish Center for Artificial Intelligence
This paper explores how robots can learn manipulation skills involving contact and friction. The authors propose a combination of Dynamic Movement Primitives (DMP) and residual learning, which significantly improves the overall performance of DMPs. The evaluation results demonstrate the effectiveness of the proposed framework.
Manipulation skills involving contact and friction are inherent to many robotics tasks. Using the class of motor primitives for peg-in-hole like insertions, we study how robots can learn such skills. Dynamic Movement Primitives (DMP) are a popular way of extracting such policies through behaviour cloning (BC) but can struggle in the context of insertion. Policy adaptation strategies such as residual learning can help improve the overall performance of policies in the context of contact-rich manipulation. However, it is not clear how to best do this with DMPs. As a result, we consider several possible ways for adapting a DMP formulation and propose residual Learning from Demonstration (rLfD), a framework that combines DMPs with Reinforcement Learning (RL) to learn a residual correction policy. Our evaluations suggest that applying residual learning directly in task space and operating on the full pose of the robot can significantly improve the overall performance of DMPs. We show that rLfD offers a gentle to the joints solution that improves the task success and generalisation of DMPs and enables transfer to different geometries and frictions through few-shot task adaptation. The proposed framework is evaluated on a set of tasks. A simulated robot and a physical robot have to successfully insert pegs, gears and plugs into their respective sockets.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据