Multi Objective Optimization For Speed And Stability Of A Sony Aibo Gait

Multi-objective Optimization for Speed and Stability of a Sony AIBO Gait

RoboCup 2006: Robot Soccer World Cup X

This book constitutes the 10th official archival publication devoted to RoboCup. It documents the achievements presented at the RoboCup 2006 International Symposium, held in Bremen, Germany, in June 2006, in conjunction with the RoboCup Competition. It serves as a valuable source of reference and inspiration for those interested in robotics or distributed intelligence.

Applying the Policy Gradient Algorithm to Learn Stable Robot Locomotion

In the RoboSoccer domain, a fast gait is an important component of a successful team. A robot with a fast gait often has body movements that cause unsteady camera motions which degrades the robot's performance during a game. This paper presents an implementation of the policy gradient machine learning algorithm that searches for a parameterized walk while optimizing for both speed and stability. The experiments were performed on the Sony Aibo ERS-7 robot platform. The resulting gait is reasonably fast and considerably more stable compared to AustinVilla's fast gait.