Login / Signup

Fast and Stable Learning of Quasi-Passive Dynamic Walking by an Unstable Biped Robot based on Off-Policy Natural Actor-Critic.

Tsuyoshi UenoYutaka NakamuraTakashi TakumaTomohiro ShibataKoh HosodaShin Ishii
Published in: IROS (2006)
Keyphrases
  • learning algorithm
  • biped robot
  • reinforcement learning
  • learning tasks
  • neural network
  • decision making
  • supervised learning
  • dynamic environments
  • degrees of freedom
  • policy gradient methods