Parareal with a learned coarse model for robotic manipulation

Parareal with a learned coarse model for robotic manipulation
复制标题

Parareal 具有用于机器人操作的学习粗略模型

DOI:
10.1007/s00791-020-00327-0
复制
发表时间:
2020
影响因子:
--
通讯作者:
Agboh W
Agboh W
中科院分区:
--
文献类型:
--
作者:
Agboh W

文献摘要

参考文献

被引文献

相似文献

许多机器人基于模型的规划和控制算法的关键组成部分是物理预测,即在给定初始状态和控制序列的情况下预测状态序列。这个过程是缓慢的,并且是机器人规划算法的主要计算瓶颈。时间积分方法可以帮助利用并行计算来加速物理预测和规划。Parareal算法在粗串行积分器和细并行积分器之间迭代。一个关键的挑战是设计一个粗略的模型,计算成本低,但足够准确,Parareal快速收敛。在这里,我们研究了在机器人操作的背景下使用深度神经网络物理模型作为Parareal的粗略模型。在使用物理引擎Mujoco作为精细传播器的模拟实验中,我们表明,学习的粗模型比基于物理的粗模型更快地Parareal收敛。我们进一步表明,学习的粗糙模型允许将Parareal应用于具有多个对象的场景,其中基于物理的粗糙模型不适用。最后,我们进行了实验,一个真实的机器人,并表明Parareal预测接近现实世界的物理预测机器人推动多个对象。代码( https://doi.org/10.5281/zenodo.3779085 )和视频( https://youtu.be/wCh2o1rf-gA )是公开的。
A key component of many robotics model-based planning and control algorithms is physics predictions, that is, forecasting a sequence of states given an initial state and a sequence of controls. This process is slow and a major computational bottleneck for robotics planning algorithms. Parallel-in-time integration methods can help to leverage parallel computing to accelerate physics predictions and thus planning. The Parareal algorithm iterates between a coarse serial integrator and a fine parallel integrator. A key challenge is to devise a coarse model that is computationally cheap but accurate enough for Parareal to converge quickly. Here, we investigate the use of a deep neural network physics model as a coarse model for Parareal in the context of robotic manipulation. In simulated experiments using the physics engine Mujoco as fine propagator we show that the learned coarse model leads to faster Parareal convergence than a coarse physics-based model. We further show that the learned coarse model allows to apply Parareal to scenarios with multiple objects, where the physics-based coarse model is not applicable. Finally, we conduct experiments on a real robot and show that Parareal predictions are close to real-world physics predictions for robotic pushing of multiple objects. Code ( https://doi.org/10.5281/zenodo.3779085 ) and videos ( https://youtu.be/wCh2o1rf-gA ) are publicly available.
DOI: 10.1007/978-3-030-44051-0_38
发表时间: 2018-12
期刊: --
影响因子: --
作者:
Brian Plancher;S. Kuindersma
通讯作者: Brian Plancher;S. Kuindersma
DOI: 10.1007/978-3-030-44051-0_40
发表时间: 2020
期刊: Workshop on the Algorithmic Foundations of Robotics
影响因子: --
作者:
Fan, T.;Schultz, J.;Murphey, T.
通讯作者: Murphey, T.
通过人在环在混乱中进行非预先操纵
DOI: --
发表时间: 2020
期刊: --
影响因子: --
作者:
Papallas R
通讯作者: Papallas R
快推和慢推:不确定性下非综合操纵的任务自适应规划
DOI: --
发表时间: 2018
期刊: --
影响因子: --
作者:
Agboh W
通讯作者: Agboh W
K.Sakoda:“光子晶体的光学特性”施普林格出版社。
DOI: --
发表时间: --
期刊:
影响因子: --
作者:
通讯作者: --