Data-Efficient Model Learning and Prediction for Contact-Rich Manipulation Tasks
Data-Efficient Model Learning and Prediction for Contact-Rich Manipulation Tasks
复制标题
针对接触丰富的操作任务的数据高效模型学习和预测
DOI：
--
复制
发表时间：
2020
影响因子：
5.2
通讯作者：
D. Kragic
中科院分区：
文献类型：
--
作者：
S. A. Khader;Hang Yin;P. Falco;D. Kragic
在这封信中，我们研究了对富含接触的操作的状态变量的远期动态模型（长期预测）两个方面的动态和数据效率 - 在确定的范围中很重要，并对最先进的方法构成了重大挑战。这明确适应了模型的特定混合结构，同时利用高斯过程的不确定性表示和数据效率。 。
In this letter, we investigate learning forward dynamics models and multi-step prediction of state variables (long-term prediction) for contact-rich manipulation. The problems are formulated in the context of model-based reinforcement learning (MBRL). We focus on two aspects–discontinuous dynamics and data-efficiency–both of which are important in the identified scope and pose significant challenges to State-of-the-Art methods. We contribute to closing this gap by proposing a method that explicitly adopts a specific hybrid structure for the model while leveraging the uncertainty representation and data-efficiency of Gaussian process. Our experiments on an illustrative moving block task and a 7-DOF robot demonstrate a clear advantage when compared to popular baselines in low data regimes.
非线性分段平滑混合系统的无监督学习
DOI：
--
发表时间：
2017
期刊：
arXiv.org
影响因子：
--
作者：
Lee, Gilwoo;Marinho, Zita;Johnson, Aaron M.;Gordon, Geoff J.;Srinivasa, Siddhartha S.;Mason, Matthew T.
通讯作者：
Mason, Matthew T.