Robotica

Table of Contents - 2004 - Volume 22, Issue 01  

As this article doesn't contain an abstract, the image below is necessary to enable the article to be indexed by certain search engines. The resolution of the full-text PDF is much higher than that shown here.
Robotica (2004), 22:1:61-75 Cambridge University Press
Copyright © 2004 Cambridge University Press
doi:10.1017/S0263574703005356

Multi-phase sumo maneuver learning

Jiming  Liu a1 and Shiwu  Zhang a2
a1 Department of Computer Science, Hong Kong Baptist University, Kowloon Tong, Hong Kong (P.R. of China). E-mail: jiming@comp.hkbu.edu.hk
a2 Department of Precision Machinery & Precision Instrumentation, University of Science and Technology of China (P.R. of China).

Article author query
liu j   [Google Scholar] 
zhang s   [Google Scholar] 
 

Abstract

In this paper, we demonstrate a multi-phase genetic programming (MPGP) approach to an autonomous robot learning task, where a sumo wrestling robot is required to execute specialized pushing maneuvers in response to different opponents' postures. The sumo robot used has a very simple, minimalist hardware configuration. This example differs from the earlier studies in evolutionary robotics in that the former is carried out on-line during the performance of a robot, whereas the latter is concerned with the evolution of a controller in a simulated environment based on extended genetic algorithms. As illustrated in several sumo maneuver learning experiments, strategic maneuvers with respect to some possible changes in the shape and size of an opponent can readily emerge from the on-line MPGP learning sessions.

(Received May 16 2003)


Key Words: Multi-phase genetic programming (MPGP); Autonomous robots; Sumo tasks; Maneuver learning; Evolutionary robotics.