Title

Confronting The Challenge Of Learning A Flexible Neural Controller For A Diversity Of Morphologies

Keywords

HyperNEAT; Legged robots; NEAT; Neuroevolution

Abstract

The ambulatory capabilities of legged robots offer the potential for access to dangerous and uneven terrain without a risk to human life. However, while machine learning has proven effective at training such robots to walk, a significant limitation of such approaches is that controllers trained for a specific robot are likely to fail when transferred to a robot with a slightly different morphology. This paper confronts this challenge with a novel strategy: Instead of training a controller for a particular quadruped morphology, it evolves a special function (through a method called Hyper-NEAT) that takes morphology as input and outputs an entire neural network controller fitted to the specific morphology. Once such a relationship is learned the output controllers are able to work on a diversity of different morphologies. Highlighting the unique potential of such an approach, in this paper a neural controller evolved for three different robot morphologies, which differ in the length of their legs, can interpolate to never-seen intermediate morphologies without any further training. Thus this work suggests a new research path towards learning controllers for whole ranges of morphologies: Instead of learning controllers themselves, it is possible to learn the relationship between morphology and control. Copyright © 2013 ACM.

Publication Date

9-2-2013

Publication Title

GECCO 2013 - Proceedings of the 2013 Genetic and Evolutionary Computation Conference

Number of Pages

255-261

Document Type

Article; Proceedings Paper

Personal Identifier

scopus

DOI Link

https://doi.org/10.1145/2463372.2463397

Socpus ID

84883116028 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/84883116028

This document is currently not available here.

Share

COinS