Estimating biped gait using spline-based probability distribution function with Q-learning

doi:10.1109/TIE.2007.908526

Return

Estimating biped gait using spline-based probability distribution function with Q-learning

2008-03-01

PRE

Lingyun Hu *

DOI：10.1109/TIE.2007.908526

Original

Original request for help

Save

Abstract

En 中文

This paper studies the probability distribution functions of the parameters to be learned and optimized in biped gait generation. By formulating the gait pattern generation into a multiobjective optimization problem with consideration of geometric and state constraints, dynamically stable and low energy cost biped gaits are generated and optimized by the proposed method, namely Spline-based Estimation of Distribution Algorithm (EDA) with Q-learning updating rule (EDA_S_Q). Instead of assuming variables as independent ones, the relationship between them is exploited by formulating the corresponding probability models with the Catmull-Rom cubic spline function. Such kind of function is proved to be a suboptimal and adaptive realization of the cubic spline function and is capable of providing high-precision description. Moreover, the probability models are updated autonomously by Q-learning method, which is model-free and adaptive. Thus, EDA_S_Q can deal with complex probability distribution functions without a prior knowledge about the distribution. The biped gait generated by EDA_S_Q has been verified using the simulation model of a humanoid soccer robot Robo-Erectus. It also shows that EDA_S_Q can generate the desired biped gaits autonomously in short learning epochs. An interpretation of the transition probability distribution achieved by EDA_S_Q provides us easy understanding for biped locomotion and better control in humanoid robots.

Keywords:

biped robot

Estimation of Distribution Algorithm (EDA)

gait pattern generation

probability model

Q-learning

spline function

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.