Combining Physical Simulators and Object-Based Networks for Control

Combining Physical Simulators and Object-Based Networks for Control
复制标题

结合物理模拟器和基于对象的网络进行控制

DOI:
10.1109/icra.2019.8794358
复制
发表时间:
2019
期刊:
2019 International Conference on Robotics and Automation (ICRA)
影响因子:
--
通讯作者:
L. Kaelbling
L. Kaelbling
中科院分区:
--
文献类型:
--
作者:
Anurag Ajay;Maria Bauzá;Jiajun Wu;Nima Fazeli;J. Tenenbaum;Alberto Rodriguez;L. Kaelbling

文献摘要

参考文献

被引文献

相似文献

物理引擎在机器人规划和控制中起着重要的作用;然而,许多现实世界的控制问题涉及复杂的接触动力学,无法解析。因此,大多数物理引擎采用近似,导致精度损失。在本文中,我们提出了一种混合动力学模型,模拟器增强的交互网络(SAIN),结合物理引擎与基于对象的神经网络的动力学建模。与现有的纯分析或纯数据驱动的模型相比,我们的混合模型以更准确和更有效的数据方式捕获交互对象的动态。在仿真和真实的机器人上的实验表明,当用于复杂的控制任务时,它也会导致更好的性能。最后,我们证明了我们的模型可以推广到具有不同物体形状和材料的新环境。
Physics engines play an important role in robot planning and control; however, many real-world control problems involve complex contact dynamics that cannot be characterized analytically. Most physics engines therefore employ approximations that lead to a loss in precision. In this paper, we propose a hybrid dynamics model, simulator-augmented interaction networks (SAIN), combining a physics engine with an object-based neural network for dynamics modeling. Compared with existing models that are purely analytical or purely data-driven, our hybrid model captures the dynamics of interacting objects in a more accurate and data-efficient manner. Experiments both in simulation and on a real robot suggest that it also leads to better performance when used in complex control tasks. Finally, we show that our model generalizes to novel environments with varying object shapes and materials.
DOI: --
发表时间: 2018-04
期刊: ArXiv
影响因子: --
作者:
A. Srinivas;A. Jabri;P. Abbeel;S. Levine;Chelsea Finn
通讯作者: A. Srinivas;A. Jabri;P. Abbeel;S. Levine;Chelsea Finn
DOI: 10.15607/rss.2017.xiii.040
发表时间: 2017-05
期刊: ArXiv
影响因子: --
作者:
Jiaji Zhou;J. Bagnell;M. T. Mason
通讯作者: Jiaji Zhou;J. Bagnell;M. T. Mason