Multi-Objective Algorithms for Learning Open-Ended Robotic Problems

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Robert, Martin, Brodeur, Simon, Ferland, Francois
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915017314533376
author Robert, Martin
Brodeur, Simon
Ferland, Francois
author_facet Robert, Martin
Brodeur, Simon
Ferland, Francois
contents Quadrupedal locomotion is a complex, open-ended problem vital to expanding autonomous vehicle reach. Traditional reinforcement learning approaches often fall short due to training instability and sample inefficiency. We propose a novel method leveraging multi-objective evolutionary algorithms as an automatic curriculum learning mechanism, which we named Multi-Objective Learning (MOL). Our approach significantly enhances the learning process by projecting velocity commands into an objective space and optimizing for both performance and diversity. Tested within the MuJoCo physics simulator, our method demonstrates superior stability and adaptability compared to baseline approaches. As such, it achieved 19\% and 44\% fewer errors against our best baseline algorithm in difficult scenarios based on a uniform and tailored evaluation respectively. This work introduces a robust framework for training quadrupedal robots, promising significant advancements in robotic locomotion and open-ended robotic problems.
format Preprint
id arxiv_https___arxiv_org_abs_2411_08070
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Multi-Objective Algorithms for Learning Open-Ended Robotic Problems
Robert, Martin
Brodeur, Simon
Ferland, Francois
Robotics
Neural and Evolutionary Computing
Quadrupedal locomotion is a complex, open-ended problem vital to expanding autonomous vehicle reach. Traditional reinforcement learning approaches often fall short due to training instability and sample inefficiency. We propose a novel method leveraging multi-objective evolutionary algorithms as an automatic curriculum learning mechanism, which we named Multi-Objective Learning (MOL). Our approach significantly enhances the learning process by projecting velocity commands into an objective space and optimizing for both performance and diversity. Tested within the MuJoCo physics simulator, our method demonstrates superior stability and adaptability compared to baseline approaches. As such, it achieved 19\% and 44\% fewer errors against our best baseline algorithm in difficult scenarios based on a uniform and tailored evaluation respectively. This work introduces a robust framework for training quadrupedal robots, promising significant advancements in robotic locomotion and open-ended robotic problems.
title Multi-Objective Algorithms for Learning Open-Ended Robotic Problems
topic Robotics
Neural and Evolutionary Computing
url https://arxiv.org/abs/2411.08070