Reinforcement Learning of Depth Stabilization with a Micro Diving Agent - UTU Research Portal

A4 Refereed article in a conference publication

Reinforcement Learning of Depth Stabilization with a Micro Diving Agent

Authors: Brinkmann Gerrit, Bessa Wallace M., Duecker Daniel A., Kreuzer Edwin, Solowjow Eugen

Editors: Kevin Lynch

Conference name: IEEE International Conference on Robotics and Automation

Publication year: 2018

Journal: IEEE International Conference on Robotics and Automation

Book title : Proceedings of the 2018 IEEE International Conference on Robotics and Automation (ICRA)

Series title: IEEE International Conference on Robotics and Automation

First page : 6197

Last page: 6203

ISBN: 978-1-5386-3082-2

eISBN: 978-1-5386-3081-5

ISSN: 2152-4092

DOI: https://doi.org/10.1109/ICRA.2018.8461137

Web address : https://ieeexplore.ieee.org/document/8461137

Abstract

Reinforcement learning (RL) allows robots to solve control tasks through interaction with their environment. In this paper we study a model-based value-function RL approach, which is suitable for computationally limited robots and light embedded systems. We develop a diving agent, which uses the RL algorithm for underwater depth stabilization. Simulations and experiments with the micro diving agent demonstrate its ability to learn the depth stabilization task.