Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Olfaction01:25

Olfaction

44.7K
The sense of smell is achieved through the activities of the olfactory system. It starts when an airborne odorant enters the nasal cavity and reaches olfactory epithelium (OE). The OE is protected by a thin layer of mucus, which also serves the purpose of dissolving more complex compounds into simpler chemical odorants. The size of the OE and the density of sensory neurons varies among species; in humans, the OE is only about 9-10 cm2.
The olfactory receptors are embedded in the cilia of the...
44.7K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Publisher Correction: A critical assessment of reinforcement learning methods for microswimmer navigation in complex flows.

The European physical journal. E, Soft matter·2025
Same author

A critical assessment of reinforcement learning methods for microswimmer navigation in complex flows.

The European physical journal. E, Soft matter·2025
Same author

The nonlinear motion of cells subject to external forces.

Soft matter·2022
See all related articles

Related Experiment Video

Updated: Aug 6, 2025

An Objective and Reproducible Test of Olfactory Learning and Discrimination in Mice
09:33

An Objective and Reproducible Test of Olfactory Learning and Discrimination in Mice

Published on: March 22, 2018

8.7K

Deep reinforcement learning for the olfactory search POMDP: a quantitative benchmark.

Aurore Loisy1, Robin A Heinonen2

  • 1Aix Marseille Univ, CNRS, Centrale Marseille, IRPHE, Marseille, France. aurore.loisy@irphe.univ-mrs.fr.

The European Physical Journal. E, Soft Matter
|March 20, 2023
PubMed
Summary

Deep reinforcement learning offers a competitive approach for olfactory search problems, mimicking insect odor tracking. This method generates efficient policies for robots, outperforming traditional solvers in computational cost and suitability.

More Related Videos

A Lateralized Odor Learning Model in Neonatal Rats for Dissecting Neural Circuitry Underpinning Memory Formation
10:42

A Lateralized Odor Learning Model in Neonatal Rats for Dissecting Neural Circuitry Underpinning Memory Formation

Published on: August 18, 2014

9.0K
Controlled Odor Mimic Permeation Systems for Olfactory Training and Field Testing
05:54

Controlled Odor Mimic Permeation Systems for Olfactory Training and Field Testing

Published on: January 28, 2021

4.7K

Related Experiment Videos

Last Updated: Aug 6, 2025

An Objective and Reproducible Test of Olfactory Learning and Discrimination in Mice
09:33

An Objective and Reproducible Test of Olfactory Learning and Discrimination in Mice

Published on: March 22, 2018

8.7K
A Lateralized Odor Learning Model in Neonatal Rats for Dissecting Neural Circuitry Underpinning Memory Formation
10:42

A Lateralized Odor Learning Model in Neonatal Rats for Dissecting Neural Circuitry Underpinning Memory Formation

Published on: August 18, 2014

9.0K
Controlled Odor Mimic Permeation Systems for Olfactory Training and Field Testing
05:54

Controlled Odor Mimic Permeation Systems for Olfactory Training and Field Testing

Published on: January 28, 2021

4.7K

Area of Science:

  • Robotics and Artificial Intelligence
  • Computational Neuroscience
  • Environmental Monitoring

Background:

  • The olfactory search problem, modeled as a partially observable Markov decision process (POMDP), simulates insect odor-finding behavior in turbulent environments.
  • Developing efficient solutions for POMDPs is crucial for applications like autonomous navigation and environmental sensing using robots.
  • Exact solutions for complex POMDPs are computationally intractable, necessitating the exploration of approximate methods.

Purpose of the Study:

  • To quantitatively benchmark a deep reinforcement learning (DRL) based solver against traditional approximate POMDP solvers for olfactory search tasks.
  • To evaluate the effectiveness and efficiency of DRL in generating policies for odor-source localization.
  • To determine the suitability of DRL-generated policies for resource-constrained robotic platforms.

Main Methods:

  • A deep reinforcement learning approach was implemented to solve the olfactory search POMDP.
  • The DRL solver was quantitatively benchmarked against established approximate POMDP solvers.
  • Performance was evaluated based on solution quality and computational efficiency, focusing on policy generation for robotic deployment.

Main Results:

  • Deep reinforcement learning demonstrated competitive performance compared to traditional POMDP approximate solvers.
  • DRL-based methods proved effective in generating lightweight and computationally efficient policies.
  • The study highlights DRL as a viable alternative for solving complex sequential decision-making problems in robotics.

Conclusions:

  • Deep reinforcement learning presents a powerful and efficient alternative for addressing the olfactory search POMDP.
  • The generated policies are suitable for deployment on lightweight robotic systems, enabling effective odor-source localization.
  • This research advances the application of AI in robotics for tasks requiring navigation and sensing in complex environments.