UC BERKELEY
EECS technical reports
TECHNICAL REPORTS


CSD-05-1395.pdf
Conditions of Use

Archive Home Page

Learning to Move Autonomously in a Hostile World

Authors:
Ikemoto, Leslie
Arikan, Okan
Forsyth, David
Technical Report Identifier: CSD-05-1395
June 2005
CSD-05-1395.pdf

Abstract: This paper describes a framework for controlling autonomous agents. We estimate an optimal controller using a novel reinforcement learning method based on stochastic optimization. The agent's skeletal configurations are taken from a motion graph which contains seamless transitions, guaranteeing smooth, natural-looking motion. The controller learns a parametric value function for choosing transitions at the branch points in the motion graph. Since this query can be completed quickly, synthesis is performed online in real-time. We couple the local controller with a global path planner to create a system which produces realistic motion even in a rapidly changing environment.