Faculty Advisor

Brown, David C.

Abstract

Our project used decision tree learning in combination with layered learning to teach agents soccer skills within the Robocup simulated soccer environment. We examined the numerous factors affecting the agent's ability to learn. We successfully trained the agent to perform two lower level skills, passing and shooting. Then we had similar success training the agent to use an upper level skill that decides between the two lower level skills. We also identified parameters that allowed for maximum learning.

Publisher

Worcester Polytechnic Institute

Date Accepted

January 2002

Major

Computer Science

Project Type

Major Qualifying Project

Accessibility

Restricted-WPI community only

Advisor Department

Computer Science

Advisor Program

Computer Science

Share

COinS