Reinforcement learning of walking behavior for a four-legged robot

Reinforcement learning of walking behavior for a four-legged robot
复制标题

四足机器人行走行为的强化学习

DOI:
10.1109/cdc.2001.980135
复制
发表时间:
2001
期刊:
Proceedings of the 40th IEEE Conference on Decision and Control (Cat. No.01CH37228)
影响因子:
--
通讯作者:
S. Kobayashi
S. Kobayashi
中科院分区:
--
文献类型:
--
作者:
H. Kimura;Toru Yamashita;S. Kobayashi

文献摘要

参考文献

被引文献

相似文献

本文研究了四足机器人行走行为的强化学习问题。机器人每条腿有两个伺服电机,因此该问题具有八维连续状态/动作空间。提出了一种适用于行动者-批评者算法的动作选择方案,在该方案中,参与者使用正态分布从其有界的动作空间中选择一个连续的动作。实验结果表明,在实际的学习步骤中,机器人成功地学会了行走。
In this paper, we investigate a reinforcement learning of walking behavior for a four-legged robot. The robot has two servo motors per leg, so this problem has eight-dimensional continuous state/action space. We present an action selection scheme for actor-critic algorithms, in which the actor selects a continuous action from its bounded action space by using the normal distribution. The experimental results show the robot successfully learns to walk in practical learning steps.
DOI: 10.1177/105971239700600201
发表时间: 1997-09
期刊: Adaptive Behavior
影响因子: 1.6
作者:
J. Santamaría;R. Sutton;A. Ram
通讯作者: J. Santamaría;R. Sutton;A. Ram