Home /Research /Evolutionary Approaches to Walking and Higher-Level Behaviors in 6-Legged Animats
LOCOMOTION

Evolutionary Approaches to Walking and Higher-Level Behaviors in 6-Legged Animats

Jean-Arcady Meyer

Year
1998
Citations
11

Abstract

This article describes the main current research project in evolutionary robotics at the AnimatLab, Paris. It aims at using an artificial selection process to automatically generate neural networks that control walking animats, i.e., simulated insects or real legged-robots. Essentially, it complements an underlying evolutionary process with a developmental procedure -- in order to reduce the size of the genotypic space that is explored -- and it calls upon an incremental approach -- in order to capitalize upon previously found solutions to simpler problems for solving problems of increasing difficulties. This

Keywords

Computer scienceArtificial intelligenceEvolutionary roboticsProcess (computing)RobotEvolutionary algorithmArtificial neural networkArtificial lifeRoboticsSelection (genetic algorithm)

Related papers

Browse all LOCOMOTION papers