Physics > Fluid Dynamics
[Submitted on 23 Aug 2018 (this version), latest version 4 Dec 2018 (v2)]
Title:Self-learning how to swim at low Reynolds number
View PDFAbstract:Synthetic microswimmers show great promise in biomedical applications such as drug delivery and microsurgery. However, their locomotion is subject not only to stringent constraints due to physical laws at microscales but also to uncontrolled environmental factors in realistic biological media. Successful applications of these synthetics are contingent upon their ability to adapt their locomotory gaits across varying biological environments. Here, we present a machine learning framework to design a new class of self-learning, adaptive (or "smart") swimmers at low Reynolds numbers. Unlike the conventional approach of designing synthetic microswimmers, we do not specify any locomotory gaits $\textit{a priori}$ but allow the swimmer to self-learn its own propulsion policy based on its interactions with the surrounding medium via reinforcement learning. We showcase the capabilities of these smart swimmers to identify effective propulsion policies, progressively improve these policies, and adapt their locomotory gaits to traverse media with vastly different properties. Further, these swimmers perform robustly under the influence of random noises, and the learning algorithm is scalable to complex designs with multiple degrees of freedom. We demonstrate these novel features theoretically via a simple reconfigurable system amenable to future experimental implementation. Our studies lay the groundwork for designing the next-generation of smart micro-robots with robust locomotive capabilities.
Submission history
From: On Shun Pak [view email][v1] Thu, 23 Aug 2018 05:37:19 UTC (1,703 KB)
[v2] Tue, 4 Dec 2018 23:47:31 UTC (1,668 KB)
Current browse context:
physics.flu-dyn
Change to browse by:
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.