Home > Papers from 2015 > The generality of navigation by familiarity

The generality of navigation by familiarity

The insect navigation literature has led directly to simple navigation algorithms. For instance, models of ant route navigation can be based on a set of views stored following the learning of a complex route. At points during recapitulation, these views can be used to determine which direction looks the most familiar, with this simple process driving navigation. The benefit of algorithms like these is that the agent doesn’t need a complex highly structured memory and learning of routes or places can be built on the same principles. Here we investigate whether algorithms of this class can work in visual environments quite different from the panoramic, low resolution world experienced by ants. We simulated an flying agent with a narrow downwards facing sensor by tiling satellite images from Google maps. We were then able to simulate training runs and recapitulation flights. We find that simple navigation algorithms are effective in this different visual ecology, even with simple sensors.
Gaffin, D. D., Dewar, A., Graham, P., & Philippides, A. (2015). Insect-Inspired Navigation Algorithm for an Aerial Agent Using Satellite Imagery. PloS one,10(4).
Categories: Papers from 2015
  1. No comments yet.
  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s