Bayesian uncertainty quantification for micro-swimmers with fully resolved hydrodynamics

TL;DR

This paper introduces a parallelizable Bayesian framework for uncertainty quantification in complex micro-swimmer models with fully resolved hydrodynamics, enabling efficient parameter estimation from noisy data.

Contribution

It presents a novel, highly parallelizable Bayesian method that makes parameter estimation for complex hydrodynamic micro-swimmer models computationally feasible.

Findings

Demonstrated robustness in estimating fluid and elastic parameters from noisy data.

Showed potential for applying the methodology to real micro-swimmer data.

Facilitated better understanding of micro-swimmer motility patterns.

Abstract

Due to the computational complexity of micro-swimmer models with fully resolved hydrodynamics, parameter estimation has been prohibitively expensive. Here, we describe a Bayesian uncertainty quantification framework that is highly parallelizable, making parameter estimation for complex forward models tractable. Using noisy in-silico data for swimmers, we demonstrate the methodology's robustness in estimating the fluid and elastic swimmer parameters. Our proposed methodology allows for analysis of real data and demonstrates potential for parameter estimation for various types of micro-swimmers. Better understanding the movement of elastic micro-structures in a viscous fluid could aid in developing artificial micro-swimmers for bio-medical applications as well as gain a fundamental understanding of the range of parameters that allow for certain motility patterns.