Modular Reinforcement Learning In Long-Horizon Manipulation Tasks

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21730%2F24%3A00377216" target="_blank" >RIV/68407700:21730/24:00377216 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1007/978-3-031-72359-9_22" target="_blank" >https://doi.org/10.1007/978-3-031-72359-9_22</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-031-72359-9_22" target="_blank" >10.1007/978-3-031-72359-9_22</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Modular Reinforcement Learning In Long-Horizon Manipulation Tasks

Popis výsledku v původním jazyce

Recently, a number of reinforcement learning (RL) algorithms have been proposed in the area of robotic manipulation. As most of the current robotic benchmarks are focused on simple, non-diverse tasks such as the translation of objects within the scene, various singlepolicy algorithms are able to solve them with a high success rate. However, when a sequence of diverse subgoals is required (translation, rotation, 6DOF manipulation, trajectory following), the single-policy networks are shown to fail. In this work, we propose two modular multipolicy algorithms (MultiPPO2 and MultiACKTR) that improve diverse long-horizon tasks by adopting a separate policy for each skill that follows its own subgoal. We tested our algorithm in a virtual robotic simulator both on single and multi-step tasks requiring non-diverse (translation) skills and also diverse (translation, rotation and path following) skills. Both algorithms (MultiPPO2 and MultiACKTR) achieved similar performance as single-policy algorithms in the single-ste

Název v anglickém jazyce

Modular Reinforcement Learning In Long-Horizon Manipulation Tasks

Popis výsledku anglicky

Recently, a number of reinforcement learning (RL) algorithms have been proposed in the area of robotic manipulation. As most of the current robotic benchmarks are focused on simple, non-diverse tasks such as the translation of objects within the scene, various singlepolicy algorithms are able to solve them with a high success rate. However, when a sequence of diverse subgoals is required (translation, rotation, 6DOF manipulation, trajectory following), the single-policy networks are shown to fail. In this work, we propose two modular multipolicy algorithms (MultiPPO2 and MultiACKTR) that improve diverse long-horizon tasks by adopting a separate policy for each skill that follows its own subgoal. We tested our algorithm in a virtual robotic simulator both on single and multi-step tasks requiring non-diverse (translation) skills and also diverse (translation, rotation and path following) skills. Both algorithms (MultiPPO2 and MultiACKTR) achieved similar performance as single-policy algorithms in the single-ste

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Artificial Neural Networks and Machine Learning – ICANN 2024 33rd International Conference on Artificial Neural Networks, Lugano, Switzerland, September 17–20, 2024, Proceedings, Part IX

ISBN

978-3-031-72356-8

ISSN

0302-9743

e-ISSN

1611-3349

Počet stran výsledku

14

Strana od-do

299-312

Název nakladatele

Springer, Cham

Místo vydání

—

Místo konání akce

Lugano-Viganello

Datum konání akce

17. 9. 2024

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

001331898500022