Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Dong, Beltrame, Giovanni
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2407.00290
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916307459375104
author Wang, Dong
Beltrame, Giovanni
author_facet Wang, Dong
Beltrame, Giovanni
contents Traditional reinforcement learning (RL) generates discrete control policies, assigning one action per cycle. These policies are usually implemented as in a fixed-frequency control loop. This rigidity presents challenges as optimal control frequency is task-dependent; suboptimal frequencies increase computational demands and reduce exploration efficiency. Variable Time Step Reinforcement Learning (VTS-RL) addresses these issues with adaptive control frequencies, executing actions only when necessary, thus reducing computational load and extending the action space to include action durations. In this paper we introduce the Multi-Objective Soft Elastic Actor-Critic (MOSEAC) method to perform VTS-RL, validating it through theoretical analysis and experimentation in simulation and on real robots. Results show faster convergence, better training results, and reduced energy consumption with respect to other variable- or fixed-frequency approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2407_00290
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Variable Time Step Reinforcement Learning for Robotic Applications
Wang, Dong
Beltrame, Giovanni
Robotics
Traditional reinforcement learning (RL) generates discrete control policies, assigning one action per cycle. These policies are usually implemented as in a fixed-frequency control loop. This rigidity presents challenges as optimal control frequency is task-dependent; suboptimal frequencies increase computational demands and reduce exploration efficiency. Variable Time Step Reinforcement Learning (VTS-RL) addresses these issues with adaptive control frequencies, executing actions only when necessary, thus reducing computational load and extending the action space to include action durations. In this paper we introduce the Multi-Objective Soft Elastic Actor-Critic (MOSEAC) method to perform VTS-RL, validating it through theoretical analysis and experimentation in simulation and on real robots. Results show faster convergence, better training results, and reduced energy consumption with respect to other variable- or fixed-frequency approaches.
title Variable Time Step Reinforcement Learning for Robotic Applications
topic Robotics
url https://arxiv.org/abs/2407.00290