Steer ‘microswimmers’ through choppy waters
(Nanowerk News) New research looks at navigation strategies for deformable micro-swimmers in viscous fluids exposed to currents, strain, and other deformations.
A shape-shifting microswimmer is a small-scale organism or artificial structure that uses sinusoidal body waves to propel itself through a liquid environment.
The term applies to organisms such as bacteria that navigate through fluids using whip-like tails called flagella, sperm cells propel themselves through the female reproductive system, and even nematodes, tiny worms that move through water or soil by waves. Microswimmers may also describe small microrobots fabricated from soft materials designed to respond to stimuli and perform tasks such as drug delivery on a microscale.
That means the study of microswimmers has applications in a wide variety of scientific fields, from biology to basic physics to nanorobots.
In a new paper at EPJ D (“Directing corrugated micro swimmers in fluid flow through reinforcement learning”) by Jérémie Bec, a researcher at the CNRS and the Center Inria d’Université Côte d’Azur and colleagues looking to find an optimal navigation policy for micro-swimmers, which is important for increasing their performance, functionality and versatility for applications such as targeted drug delivery.
“Finding optimal navigation policies for micro swimmers is critical to increasing their performance, functionality and versatility in the applications mentioned,” said Bec. “By determining optimal navigation policies, micro-swimmers can effectively adapt and respond to changing fluid environments. This allows them to navigate through obstacles, avoid hazards, and exploit flow patterns for better locomotion.
“An optimal navigation policy ensures their ability to maneuver and navigate their surroundings efficiently,” added Bec.
Researchers explain that further than that, an optimal navigation policy guarantees robust performance across a wide range of conditions and variations as they wave through changing environments.
Bec said the team was particularly intrigued by the marked degree of variability in the performance of the machine learning strategies they used. Unexpected variability in performance provides teams with valuable insights and allows them to identify optimal strategies that exceed their initial expectations.
“We gain a deeper understanding of the complex dynamics involved in optimizing navigation policies for micro swimmers,” Bec concludes. “These findings underscore the importance of exploring beyond conventional expectations and embracing the potential for variability and uncertainty in artificial intelligence.”
