Reinforcement learning with function approximation for 3-spheres swimmer - CeMoSiS
Preprints, Working Papers, ... Year : 2022

Reinforcement learning with function approximation for 3-spheres swimmer

Abstract

We study the swimming strategies that maximize the speed of the three-sphere swimmer using reinforcement learning methods. First of all, we ensure that for a simple model with few actions, the Q-learning method converges. However, this latter method does not fit a more complex framework (for instance the presence of boundary) where states or actions have to be continuous to obtain all directions in the swimmer's reachable set. To overcome this issue, we investigate another method from reinforcement learning which uses function approximation, and benchmark its results in absence of walls.
Fichier principal
Vignette du fichier
Article_IFAC_CAO_2021-4.pdf (409.42 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03538754 , version 1 (21-01-2022)

Identifiers

  • HAL Id : hal-03538754 , version 1

Cite

Luca Berti, Zakarya El Khiyati, Youssef Essousy, Christophe Prud'Homme, Laetitia Giraldi. Reinforcement learning with function approximation for 3-spheres swimmer. 2022. ⟨hal-03538754⟩
320 View
149 Download

Share

More