A Comparative Study among Three Automatic Gait Generation Methods for Quadruped Robots

Kisung SEO  Soohwan HYUN  

IEICE TRANSACTIONS on Information and Systems   Vol.E97-D   No.2   pp.353-356
Publication Date: 2014/02/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.E97.D.353
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Artificial Intelligence, Data Mining
robot automatic gait generation,  quadruped robot,  genetic algorithm,  paw trajectory,  joint trajectory,  genetic programming,  central pattern generator,  

Full Text: PDF>>
Buy this Article

This paper introduces a comparison of three automatic gait generation methods for quadruped robots: GA (Genetic Algorithm), GP (genetic programming) and CPG (Central Pattern Generator). It aims to provide a useful guideline for the selection of gait generation methods. GA-based approaches seek to optimize paw locus in Cartesian space. GP-based techniques generate joint trajectories using regression polynomials. The CPGs are neural circuits that generate oscillatory output from an input coming from the brain. Optimizations for the three proposed methods are executed and analyzed using a Webots simulation of the quadruped robot built by Bioloid. The experimental comparisons and analyses provided herein will be an informative guidance for research of gait generation method.