Sunday, April 23, 2017

Honeybees to Drones! Interview with Mark Roper on visual detectors for navigation!!

You must have all experienced a honeybee buzz around you, avoiding all your attempts at hitting it. These tiny animals are amazing at navigation. They move around like superman finding little flowers with ease. What makes them so good at navigation?

Mark and colleagues wanted to understand the amazing navigation capabilities of honeybees and generate a model of it. One would imagine that the system would be highly complex, with tons of interactions. But they show that the system can not only be modeled on a very few components, but it has a simplistic architecture. This generates a small and efficient system, which can be useful for helping us generate better navigation algorithm for daily life. As an instance, drones require large amounts of battery power for flying, which restricts the distance they can travel. But using honeybees based navigation system can allow them to be lighter and less power consuming, allowing them to fly much longer distances. Please listen to Mark to know more.



To know more, please refer to:
Insect Bio-inspired Neural Network Provides New Evidence on How Simple Feature Detectors Can Enable Complex Visual Generalization and Stimulus Location Invariance in the Miniature Brain of Honeybees
Roper et al., PLoS Computational Biology, 2017

No comments:

Post a Comment