Salvato in:
Dettagli Bibliografici
Autori principali: Thibault, William, Rajendran, Vidyasagar, Melek, William, Mombaur, Katja
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2409.07846
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913498548666368
author Thibault, William
Rajendran, Vidyasagar
Melek, William
Mombaur, Katja
author_facet Thibault, William
Rajendran, Vidyasagar
Melek, William
Mombaur, Katja
contents Learning-based methods have proven useful at generating complex motions for robots, including humanoids. Reinforcement learning (RL) has been used to learn locomotion policies, some of which leverage a periodic reward formulation. This work extends the periodic reward formulation of locomotion to skateboarding for the REEM-C robot. Brax/MJX is used to implement the RL problem to achieve fast training. Initial results in simulation are presented with hardware experiments in progress.
format Preprint
id arxiv_https___arxiv_org_abs_2409_07846
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Learning Skateboarding for Humanoid Robots through Massively Parallel Reinforcement Learning
Thibault, William
Rajendran, Vidyasagar
Melek, William
Mombaur, Katja
Robotics
Learning-based methods have proven useful at generating complex motions for robots, including humanoids. Reinforcement learning (RL) has been used to learn locomotion policies, some of which leverage a periodic reward formulation. This work extends the periodic reward formulation of locomotion to skateboarding for the REEM-C robot. Brax/MJX is used to implement the RL problem to achieve fast training. Initial results in simulation are presented with hardware experiments in progress.
title Learning Skateboarding for Humanoid Robots through Massively Parallel Reinforcement Learning
topic Robotics
url https://arxiv.org/abs/2409.07846