Physical Review Journal Club: Optimal Trajectories for Bayesian Olfactory Search in Turbulent Flows: The Low Information Limit and Beyond
Wednesday, June 4, 2025 at 11 a.m ET
Watch on YouTube
Author Robin Heinonen (University of Genova) will Zoom with the Physical Review Journal Club on June 4 at 11 a.m. ET, to give a brief presentation followed by Q&A. The Q&A will be moderated by Physical Review Fluids Associate Editor, Emmanuel Villermaux (Aix-Marseille Université).
Registration is free and a video recording will be provided to all registrants.
Abstract
How do organisms, or algorithms, track down the source of a faint odor or signal in a chaotic, windy environment? In their paper, Heinonen and co-authors explore the challenge of source-seeking in turbulent flows, where information comes only from rare and random encounters with a passive tracer. Using high-fidelity numerical simulations, they investigate quasi-optimal search strategies that minimize the average time to locate the source. They analyze the structure of these strategies and compare them to the well-known infotaxis heuristic. In the presence of a strong mean wind, the optimal behavior in the absence of a detection resembles biological “casting”—a zigzag search pattern followed by a return to the origin. This motion leads to a characteristic square-root scaling of displacement over time. The authors also provide a theoretical estimate for how long searches may take in the limit of very low detection probabilities, showing agreement with Monte Carlo simulations. This work offers insights into both natural behaviors and the design of robotic or algorithmic search systems in uncertain environments.
Read Ahead
Optimal trajectories for Bayesian olfactory search in turbulent flows: The low information limit and beyond
