DCOB: Action space for reinforcement learning of high DoF robots

Akihiko Yamaguchi, Jun Takamatsu, Tsukasa Ogasawara

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Reinforcement learning (RL) for robot control is an important technology for future robots since it enables us to design a robot's behavior using the reward function. However, RL for high degree-of-freedom robot control is still an open issue. This paper proposes a discrete action space DCOB which is generated from the basis functions (BFs) given to approximate a value function. The remarkable feature is that, by reducing the number of BFs to enable the robot to learn quickly the value function, the size of DCOB is also reduced, which improves the learning speed. In addition, a method WF-DCOB is proposed to enhance the performance, where wire-fitting is utilized to search for continuous actions around each discrete action of DCOB. We apply the proposed methods to motion learning tasks of a simulated humanoid robot and a real spider robot. The experimental results demonstrate outstanding performance.

Original languageEnglish
Pages (from-to)327-346
Number of pages20
JournalAutonomous Robots
Volume34
Issue number4
DOIs
Publication statusPublished - 2013 May 1
Externally publishedYes

Keywords

  • Action space
  • Crawling
  • Humanoid robot
  • Motion learning
  • Reinforcement learning

ASJC Scopus subject areas

  • Artificial Intelligence

Fingerprint Dive into the research topics of 'DCOB: Action space for reinforcement learning of high DoF robots'. Together they form a unique fingerprint.

  • Cite this