Efficient collective swimming by harnessing vortices through deep reinforcement learning
METADATA ONLY
Loading...
Author / Producer
Date
2018-06-05
Publication Type
Journal Article
ETH Bibliography
yes
Citations
Altmetric
METADATA ONLY
Data
Rights / License
Abstract
Can fish reduce their energy expenditure by schooling? We answer affirmatively this longstanding question by combining state-of-the-art direct numerical simulations of the 3D Navier–Stokes equations with reinforcement learning, using recurrent neural networks with long short-term memory cells to account for the unsteadiness of the flow field. Surprisingly, we find that swimming behind a leader is not always associated with energetic benefits for the follower. In turn, we demonstrate that fish can improve their sustained propulsive efficiency by placing themselves at appropriate locations in the wake of other swimmers and intercepting their wake vortices judiciously. The results show that autonomous, “smart” swimmers may exploit unsteady flow fields to reap substantial energetic benefits and have promising implications for robotic swarms.
Permanent link
Publication status
published
External links
Editor
Book title
Volume
115 (23)
Pages / Article No.
5849 - 5854
Publisher
National Academy of Sciences
Event
Edition / version
Methods
Software
Geographic location
Date collected
Date created
Subject
fish schooling; deep reinforcement learning; Autonomous navigation; energy harvesting; Recurrent neural networks
Organisational unit
03499 - Koumoutsakos, Petros (ehemalig) / Koumoutsakos, Petros (former)
02803 - Collegium Helveticum / Collegium Helveticum