1. Open-source computational simulation of moth-inspired navigation algorithm: A benchmark framework
- Author
-
Roi Gurka, Alex Liberzon, Ally R. Harari, Noam Benelli, Gregory Zilman, and Yiftach Golov
- Subjects
`Mothpy 0.0.1' - an open-source moth-inspired navigator simulator ,Computer science ,Science ,Clinical Biochemistry ,Scalar (physics) ,Simulation framework ,Benchmarking ,Method Article ,Moths ,Odor localization ,Task (project management) ,Medical Laboratory Technology ,Open source ,Feature (computer vision) ,Navigation strategies ,Benchmark (computing) ,Bio-inspiration ,Closing (morphology) ,Olfactory navigation ,Algorithm - Abstract
Olfactory navigation is defined as a task of a self-propelled navigator with some sensors capabilities to detect odor (or scalar concentration) convected and diffused in a windy environment. Known for their expertise in locating an odor source, male moths feature a bio-inspirational model of olfactory navigation using chemosensory. Many studies have developed moths-inspired algorithms based on proposed strategies of odor-sourcing. However, comparing among various bio-inspired strategies is challenging, due to the lack of a componential framework that allows statistical comparison of their performances, in a controlled environment. This work aims at closing this gap, using an open source, freely accessible simulation framework. To demonstrate the applicability of our simulated framework as a benchmarking tool, we implemented two different moth-inspired navigation strategies; for each strategy, specific modifications in the navigation module were carried out, resulting in four different navigation models. We tested the performance of moth-like navigators of these models through various wind and odor spread parameters in a virtual turbulent environment. The performance of the navigators was comprehensively analyzed using bio-statistical tests. This benchmark-ready simulation framework could be useful for the biology-oriented, as well as engineering-oriented studies, assisting in deducing the evolutionary efficient strategies and improving self-propelled autonomous systems in complex environments.•The open-source framework `Mothpy' provides a computational platform that simulates the behavior of moth-like navigators, using two main inputs to be modified by the user: (1) flow condition; and (2) navigation strategy.•`Mothpy' can be used as a benchmarking platform to compare the performance of multiple moth-like navigators, under various physical environments, and different searching strategies.•Method name: Mothpy 0.0.1' - an open-source moth-inspired navigator simulator., Graphical abstract Image, graphical abstract
- Published
- 2021