RT-HCP: Dealing with Inference Delays and Sample Efficiency to Learn Directly on Robotic Platforms

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Asri, Zakariae El, Laiche, Ibrahim, Rambour, Clément, Sigaud, Olivier, Thome, Nicolas
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866916939830394880
author Asri, Zakariae El
Laiche, Ibrahim
Rambour, Clément
Sigaud, Olivier
Thome, Nicolas
author_facet Asri, Zakariae El
Laiche, Ibrahim
Rambour, Clément
Sigaud, Olivier
Thome, Nicolas
contents Learning a controller directly on the robot requires extreme sample efficiency. Model-based reinforcement learning (RL) methods are the most sample efficient, but they often suffer from a too long inference time to meet the robot control frequency requirements. In this paper, we address the sample efficiency and inference time challenges with two contributions. First, we define a general framework to deal with inference delays where the slow inference robot controller provides a sequence of actions to feed the control-hungry robotic platform without execution gaps. Then, we compare several RL algorithms in the light of this framework and propose RT-HCP, an algorithm that offers an excellent trade-off between performance, sample efficiency and inference time. We validate the superiority of RT-HCP with experiments where we learn a controller directly on a simple but high frequency FURUTA pendulum platform. Code: github.com/elasriz/RTHCP
format Preprint
id arxiv_https___arxiv_org_abs_2509_06714
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle RT-HCP: Dealing with Inference Delays and Sample Efficiency to Learn Directly on Robotic Platforms
Asri, Zakariae El
Laiche, Ibrahim
Rambour, Clément
Sigaud, Olivier
Thome, Nicolas
Machine Learning
Learning a controller directly on the robot requires extreme sample efficiency. Model-based reinforcement learning (RL) methods are the most sample efficient, but they often suffer from a too long inference time to meet the robot control frequency requirements. In this paper, we address the sample efficiency and inference time challenges with two contributions. First, we define a general framework to deal with inference delays where the slow inference robot controller provides a sequence of actions to feed the control-hungry robotic platform without execution gaps. Then, we compare several RL algorithms in the light of this framework and propose RT-HCP, an algorithm that offers an excellent trade-off between performance, sample efficiency and inference time. We validate the superiority of RT-HCP with experiments where we learn a controller directly on a simple but high frequency FURUTA pendulum platform. Code: github.com/elasriz/RTHCP
title RT-HCP: Dealing with Inference Delays and Sample Efficiency to Learn Directly on Robotic Platforms
topic Machine Learning
url https://arxiv.org/abs/2509.06714