Low-dimensional feature extraction for humanoid locomotion using kernel dimension reduction

Low-dimensional feature extraction for humanoid locomotion using kernel dimension reduction
复制标题

使用核降维的人形运动低维特征提取

DOI:
10.1109/robot.2008.4543621
复制
发表时间:
2008
期刊:
2008 IEEE International Conference on Robotics and Automation
影响因子:
--
通讯作者:
G. Cheng
G. Cheng
中科院分区:
--
文献类型:
--
作者:
J. Morimoto;S. Hyon;C. Atkeson;G. Cheng

文献摘要

参考文献

被引文献

相似文献

我们建议使用内核降维(KDR)提取一个低维的特征空间的人形运动任务。虽然类人机器人有许多自由度,但任务相关的特征空间可以比原始状态空间的维数小得多。我们考虑应用所提出的方法,以提高类人机器人的机车性能,使用提取的低维状态空间。为了提高机车性能,我们使用了强化学习(RL)框架。虽然RL是一种有用的非线性优化器,但通常很难将RL应用于真实的机器人系统-因为需要大量的迭代来获得合适的策略。在这项研究中,我们使用提取的低维特征空间的强化学习,使学习系统可以快速提高任务性能。核降维方法允许我们提取特征空间,即使任务相关映射是非线性的。这是一个重要的属性,以提高人形机车的性能,因为步进或行走动力学涉及高度非线性动力学。我们表明,我们可以改善步进和步行的政策,通过使用RL方法提取的特征空间,使用KDR。
We propose using the kernel dimension reduction (KDR) to extract a low-dimensional feature space for humanoid locomotion tasks. Although humanoids have many degrees of freedom, task relevant feature spaces can be much smaller than the number of dimension of the original state space. We consider an application of the proposed approach to improve the locomotive performance of humanoid robots using an extracted low-dimensional state space. To improve the locomotive performance, we use a reinforcement learning (RL) framework. While RL is a useful non-linear optimizer, it is usually difficult to apply RL to real robotic systems - due to the large number of iterations required to acquire suitable policies. In this study, we use the extracted low-dimensional feature space for RL so that the learning system can improve task performance quickly. The kernel dimension reduction method allows us to extract the feature space even if the task relevant mapping is non-linear. This is an essential property to improve humanoid locomotive performance since stepping or walking dynamics involves highly nonlinear dynamics. We show that we can improve stepping and walking policies by using a RL method on an extracted feature space by using KDR.
DOI: 10.1177/0278364907084980
发表时间: 2008-02-01
影响因子: 9.2
作者:
Endo, Gen;Morimoto, Jun;Cheng, Gordon
通讯作者: Cheng, Gordon
DOI: 10.1214/08-aos637
发表时间: 2009-08-01
影响因子: 4.5
作者:
Fukumizu, Kenji;Bach, Francis R.;Jordan, Michael I.
通讯作者: Jordan, Michael I.